1. Introduction
Disabilities and impairments significantly impact a person’s ability to perform several activities. Among these activities, the ability to drive a vehicle is fundamental to fostering persons with disabilities (PwDs)’ independence [1]. However, disabilities and impairments can hinder driving abilities, especially when they result in a loss of capacity in the upper or lower limbs. Permanent impairments to upper and lower limbs often result in the inability to drive a “standard” vehicle—i.e., any vehicle designed and manufactured for persons not characterized by any disability or impairment.
PwDs with permanent limitations or impairments to upper and lower limbs may experience difficulties in maneuvering a car using the steering wheel, shifting gears with the stick shift, or breaking and accelerating using pedals. In other words, “regular cars” may not be equipped to ensure PwDs’ driving. Moreover, a PwD’s personal aids—such as a prosthesis or orthosis—cannot ensure safe driving since they enable only a portion of the functions of the limb they replace. Therefore, cars need to be adapted to meet PwDs’ physical needs for driving safely. The adaptations may involve the vehicle mechanics and the provision of user-specific aids, depending on the driver’s health condition and impairments.
Disabilities and impairments can also appear as a consequence of accidents or conditions that affect an individual after he or she has learned how to drive a regular vehicle. Thus, the modification of a vehicle implies the drivers (with disability) relearn how to drive a car with the mechanical modifications and aids that characterize its new configuration. Before that, drivers need to become acquainted with the aids and modifications that could characterize their cars.
Nonetheless, it is possible to reconfigure a vehicle in different ways to meet a PwD’s needs. The selection of the modifications depends mainly on two factors:
The driver’s health condition dictates the types of aids, devices, and mechanical modifications that can be adopted for a vehicle.
The regulations characterizing the country in which the modifications on the vehicle are executed.
Vehicular modifications constitute a legal requirement for driving a car when disabilities or impairments characterize a driver. Moreover, it is also required that drivers with disabilities regain access to a driving license (named “special driving license” in Italy). This type of driving license is granted by a medical committee, which analyzes the candidate driver’s health status and specifies the driving aids and mechanical modifications of the vehicle; in some cases, they can also specify restrictions to the use of the vehicle, depending on the severity of the impairments.
The knowledge necessary to indicate the modifications and driving aids to be provided is codified in a set of local and European laws. The aim of this paper is to introduce a Decision Support System (DSS) leveraging such knowledge to support drivers with disability in understanding the possible modification(s) their cars could undergo; in this way, the driver can be enabled as an active part of the decision and can acquire more insights of the types and functionalities of the driving aids he or she is going to use. The knowledge relevant to the DSS is formalized relying on domain ontologies—i.e., shared and formal conceptualizations of a domain of knowledge–, developed leveraging the support of the Italian National Institute for Insurance against Accidents at Work (INAIL). In Italy, INAIL is the institution that supports PwDs in obtaining the special driving license, providing candidate drivers with essential information about the regulations, evaluating the driving skills, guiding in the selection of car configurations that support candidate drivers, and making candidate drivers test some reconfigured vehicles. The DSS can be accessed via a prototypical application, which can serve both the medical committee and the candidate drivers in understanding and selecting the modifications to be applied to the PwD’s vehicle. In principle, candidate drivers and clinical personnel should use Rip@rto DSS to discuss together the modifications proposed, understanding the set(s) best suiting the driver’s preferences, and providing PwD with more information regarding specific aids.
The remainder of this paper is organized as follows: Section 2 reviews scientific literature to identify works related to the reconfiguration of vehicles for PwDs, with the aim of identifying the main trends in such a topic. Section 3 briefly illustrates the laws and regulations characterizing the framework for the adaptation of vehicles for drivers with disabilities, focusing on the European and Italian perspectives. Section 4 describes the architecture of the DSS, with particular attention to the ontological layer, its engineering phase, and its reasoning capabilities. In the same section, the prototypical desktop application of the DSS, aimed at illustrating the general features expected from the definitive version of the DSS application, is introduced. Section 6 discusses the proposed DSS and its functionalities, highlights some limitations of the proposed approach, and underlines some research directions. Finally, the Conclusions summarize the main outcomes of this work.
2. Related Work
The problem of the reconfiguration or modification of vehicles for PwDs has been addressed in the literature mostly from the rehabilitation and design [2,3], psychological and social [4], and user experience [5] perspectives. Moreover, especially during the past three decades, many studies were dedicated to the design of driving aids for PwDs. Although these works are prior to the European Framework (see Section 3), they investigated fundamental aspects of driving and PwDs, such as driving for wheelchair users [6], the development and deployment of aids and accessories for PwDs [7], and the preliminary definitions of aids prescription guidelines [8] and drivers’ safety [9].
While these perspectives are fundamental in underlying drivers with disabilities’ needs in several fields (including the acquisition of knowledge pertaining to aids and vehicle modifications), very little is said regarding the adoption of computer-based systems to foster the satisfaction of such needs. In this section, we reviewed the scientific literature to identify works addressing the topic of DSSs in the fields of vehicular modifications for PwDs by adopting a Systematic Literature Review approach. The review methods and its results are presented and briefly discussed in the following subsection.
2.1. PRISMA Literature Review
The methodology adopted for the literature review is the Preferred Reporting Items for Systematic Reviews (PRISMA) [10]. The approach consists of three steps (article retrieval, screening, and inclusion). Thus, the research question investigated was searched in the Elsevier Scopus and Web of Science databases translated into the following query:
((vehicle modif*) AND disab*))
OR
(driver* AND disab*)
OR
(((vehicle config*) OR (car config*) OR (vehicle reconfig*) OR (car reconfig*)) AND disab*)
The databases were searched for journal articles, book chapters, and conference proceedings in English, and they were limited to the subject areas of Computer Science, Engineering, and Decision Sciences. No time constraints were set for this search, and articles “in press” were also considered—since they are included in the databases.
Figure 1 schematizes the PRISMA review process and its quantitative results.
The search in the databases retrieved 516 records, which were reduced during the screening phase to 15. The screening phase (and the inclusion phase) were conducted under the following exclusion criteria:
The paper addresses autonomous vehicles, public transportations, or any other system for a PwD mobility other than his/her car
The paper investigates wheelchairs outside of their role in driving activities
The paper concerns accident prevention or vehicle traffic monitoring
The exclusion criteria allowed us to focus on identifying solutions for PwDs devoted to assisting them in driving, adapting their vehicle, and acquiring (or re-acquiring) driving skills after an impairment or disability.
The 15 works included were reduced to 9 during the inclusion phase, in which all the articles were read.
2.2. Results
The amount of included works includes three journal articles and six conference proceedings. The selected work span from 1989 to 2024. The majority of the works is concentrated in 2023 (four papers). The full-text reading conducted in the inclusion phase allows us to classify the topic(s) addressed by each work (Table 1).
Among the included works, two conference papers ([14,18]) describe specific and preliminary aspects of the DSS proposed in this work.
2.3. Discussion on the Findings of the Literature Review
The papers are equally distributed among the three main topics: three works discuss DSSs, three papers investigate the design and development of equipment to help drivers with disability cope with a specific aspect of their disabilities, and three works illustrate simulators for training PwDs for driving and to assess their progress or residual driving abilities.
The first example of DSS dedicated to suggesting adaptive equipment for drivers with disabilities among the included works is dated 1989. Wheeler et al. [11] intended to develop a tool to assist rehabilitation professionals in determining the suitable equipment for the personalization of vehicles for persons with a physical disability. The result—Knowledge-based Adaptive Driving Equipment Tool (KADET)—consists of a prototype application evaluated by experts; however, it is not specified what types of equipment KADET can suggest. In [14], the development of a domain ontology for a DSS aimed at supporting car reconfiguration for PwDs is addressed; this work is also completed by its front-end application [18], a mixed-reality application developed to support PwDs in acquiring knowledge about the equipment (as mentioned above, these two works constitute the basis of the DSS proposed here).
Regarding the studies that address the design, development, or discussion of equipment for drivers with disabilities, the topics addressed span significantly. In [12], a particular aspect of the EU Directive 2015/653 [20] is investigated. The Directive stresses that the maximum force a PwD could exert on the vehicle’s primary controls could be adjusted; in other words, the force required to operate with the brakes and steering wheel can vary. Thus, the authors defined the steering and braking operative forces in driving as necessary for the physical assessment before in-car on-road driving assessment of drivers with physical disabilities. Salem and Boutaba [15] proposed a system to convert auditory cues into displayed alerts: the system consists of a piece of equipment thought for drivers with auditory disabilities; the system adopts data-driven techniques to enable sound recognition and then compares acquired sounds’ spectrograms to properly classify the auditory cues and convert them into visual alerts. In [16], the design of a combined brake and accelerator pedal for drivers with disabilities (affecting lower limbs) is presented. The design process results consist of a pedal capable of reducing braking and accelerating activities interference and, thus, makes it easier for a PwD to drive.
In the area of simulators for the (re-)acquisition of driving abilities, Ricci et al. [13] described an open-source simulator for training and assessment of a PwD’s driving abilities (ADRIS). ADRIS allows for setting customized features (such as adaptable driving controllers) and testing in different virtual environments and conditions, enabling physiological data acquisition. Arlati and colleagues [17] introduced VRoADS, a virtual reality-based simulator, to support clinical assessors in evaluating a PwD’s driving-related abilities. VRoADS was developed as an “upgrade proposal” of the existing simulator in INAIL, reproducing the same assessing tests (steering wheel rotation, braking, peripheral view, acoustic reactions, etc.) in a more ecological and immersive environment. To grant access to driving training and acquire data on a PwD’s progress, Jones and colleagues [19] proposed a virtual reality-based simulator prototype (DriVR). Although the aims are similar to those addressed by the other two simulators proposed, DriVR adopts different technologies.
Three main directives seem to guide researchers on this matter: the development of DSSs, the provision of a simulator capable of supporting training and assessment of PwDs, and the design and development of specific pieces of equipment that should support drivers with disability in coping with the limitations imposed by their physical conditions.
While the simulators insist on the same aims (supporting the (re-)acquisition of driving abilities and assessing them in their progress or in their residual abilities), the DSSs retrieved also share similar purposes (identify and recommend equipment, or aids, or car configurations) but for different audiences. KADET [11] is thought for clinical personnel involved in the rehabilitation of drivers with disabilities, while the other DSS ([14,18]) are devoted to PwDs. The difference between the two systems could be explained in the light of their publishing years: KADET is dated 1989, while the other is dated 2022. This could mean that at the time of developing KADET, there was no consolidated (and European) regulatory framework capable of supporting clinicians in identifying the appropriate equipment; on the contrary, after 2016, a European Directive provided supranational guidance in this regard. Differently from the two previous directives, the design and development of pieces of equipment present variegated outcomes; although the aim remains the same for all the works (support drivers in coping with their physical limitations), the impairments and necessities addressed are considerably different; thus, this makes the outcome likewise different.
The very few papers included allow us to underline a significant phenomenon. The retrieved simulators and equipment devoted to PwDs and driving, as well as the works pertaining to the DSSs, point towards the adoption of the European regulatory framework as a “standard de facto” on the topics of vehicle adaptations ([12,14,17,18]).
The temporal distribution of the paper and the case of one relevant work “older” than the others indicates that the researcher’s attention on such a topic has only started to rise recently. Nevertheless, Wheeler et al.’s work [11] clearly suggests that the issue pertaining to the identification of the equipment and devices that could be used to reconfigure a PwD car is not new. Considering the year of publication of most of the works, it may be plausible that the possibility of relying on a consolidated regulatory framework could have played a pivotal role in developing this research area. The pervasiveness of the European normative is not limited to the field of DSSs, but it also guides research on equipment and simulators. Therefore, it is safe to conclude that this regulatory framework marked a milestone for the European countries in the area of car reconfiguration for PwDs. Nonetheless, it is evident that the European laws are far from perfect, as underlined in [12]. Yet, they provide a fundamental perspective for the topic addressed.
Another noticeable fact pertaining to the DSSs in the investigated research area is the lack of data-driven solutions. This may indicate that there are no exhaustive datasets on car reconfigurations for PwDs and that the knowledge-based approach may be (so far) the only feasible way to arrive to a DSS. This would also imply that knowledge-based solutions would require reliable sources of knowledge.
2.4. Findings and Remarks
The literature review underlines that PwDs and driving is a very recent topic. However, there exists a variety of solutions to such a complex problem, and this phenomenon is reflected in the results of the review (Table 1). The very few papers related to the development of DSSs are dated 1989 and, more recently, 2022 and 2023. The latter two studies are preliminary works of the DSS described here and, thus, leaves KADET [11] as the only DSS to compare our solution with. Differently from KADET, the Rip@rto DSS leverages EU and Italian norms to formalize a semantic knowledge base; moreover, Rip@rto is devoted to the collaborative selection of aids and devices by PwDs and clinical personnel.
The review process has some limitations: the query is focused on car reconfiguration, so to elicit those works that specifically cover these particular aspects. However, it is worth observing that the evolution of research towards unmanned and autonomous vehicles may support driver with disabilities in a different way [21,22].
3. Regulatory Framework for Drivers with Disabilities and Vehicle Modifications
This section provides a few coordinates on the laws and regulations characterizing the situations of driving for PwDs and modifications of vehicles following a permanent impairment or disability. This section will focus on the case of the European Union since—as highlighted in the previous section—it seems to cover a pivotal role in the field of car reconfiguration; this is true also for the Italian context, in which INAIL operates as the institution in charge of providing PwDs with suitable car modifications.
During the past ten years, the EU has made an effort to standardize access to the special driving license, which was previously managed by single countries with different local regulations. The first effort of this homogenization effort is the EU Directive 2015/653 [20], which was adopted in all EU countries, including Italy, in the following two years. In particular, Italy incorporated this Directive in the national “Codice della strada” (a set of laws and regulations to discipline local traffic) [23].
The identification of a driver’s disabilities and impairments and the consequent selection of mechanical modifications, devices, and aids follow a strictly regulated process: the driver’s physical and cognitive abilities are assessed by specialized health professionals (a medical committee), who are in charge of identifying a person’s impairments and their magnitude and, thus, assesses his or her residual abilities for driving. The process may also result in limiting the driver’s possibility to drive under certain circumstances: for instance, a driver may be required to drive within the limits of a specified radius from his or her city, or she/he may be prevented from driving on highways, or she/he may be restricted from driving during night hours or for a time longer than a specified amount, or she/he is required to always have a passenger accompanying her/him. These specifics are coded in EU Directive 2015/653 (and in its updated version dated 2020), which provides a set of codes to identify aids to be given to drivers for medical reasons, vehicle adaptations, and administrative conditions that may characterize the driver’s license (Table 2). Italian law received these prescriptions, updating the 148/1991 Ministry of Transport circular.
Once the health condition is assessed, the regulatory framework provides a set of rules describing different modifications that can occur to a vehicle depending on its driver’s condition and impairment(s). Therefore, it is possible that one or more “car configurations” are available for a specific condition. Each configuration is composed of a set of aids and mechanical modifications to the vehicle, thought to support the PwD in coping with her/his impairments while driving. Figure 2 provides an example of the combination of impairments and possible car configurations.
The figure provides two examples of impairments that imply vehicular adaptations according to the side of the disability. The possible impairments are coded by the law (as explained in the following section) and their combinations (double impairments).
The example coded as “1PM + 1C” indicates that the PwD is characterized by one arm being replaced by a myoelectric prosthesis (1PM); in contrast, the other arm, although impaired, is still capable of performing driving-related functions.
Different is the case for impairment code “3”, which—depending on the side where the disability is located—requires the PwD to make a choice about the different options available (Figure 3).
While the process is transparent from a normative and clinical perspective, the PwDs do not necessarily have knowledge about the different configurations and the devices that may be installed in their cars. Moreover, the availability of more configurations for some conditions may require the drivers with a disability to decide which adaptations she/he may want to set up on the vehicle [5]. Therefore, it would be advisable that the PwD is familiar with the most common driving aids (and car configurations) since they significantly impact her/his driving experience.
4. Developing the Rip@rto DSS
To support drivers with disabilities in understanding the modifications their vehicles may undergo, a DSS has been developed and presented in this section. The DSS is one of the results of the “Rip@rto” research project financed by INAIL. The DSS comprises two main components: the ontology and the application acting as a Graphical User Interface (GUI) (as schematized in Figure 4).
In the following subsection, the development of the two components is described.
4.1. Ontological Layer of Rip@rto
As highlighted in Section 2, the lack of suitable datasets and the existence of regulations and norms suggested the formalization Rip@rto’s knowledge in the form of an ontology. In this way, a purely kowledge-based DSS may also foster the application acceptance, considering that “black-box” data-driven DSS may not be well tolerated by the clinicians composing the medical committee [24,25], who are ultimately legally responsible for the choices related to a PwD’s car reconfiguration. Ontological reasoning enables a transparent and human-friendly reasoning process [26], which resembles the reasoning activities performed by a human assessor [27]. For these reasons, ontologies are widely adopted in DSSs in a variety of fields [28,29,30]—particularly, in healthcare and clinical domains [31,32,33,34].
The Rip@rto ontology needs to “translate” the regulations, their codes, the possible configurations, and the combinations of configurations and impairments in formal, explicit, and shared conceptualization [35]. According to the findings of this discipline in the healthcare domain [36,37], the ontology engineering process takes advantage of both regulations and domain experts. For the development of Rip@rto, the authors had access to INAIL personnel, whose expertise in the field of vehicular modifications for PwD is precious and pivotal. Three domain experts were interviewed to gather insights on the regulations and their functioning in a real setting, and real-world scenarios were developed to validate the results of the ontology reasoning process.
The rigorous normative framework depicting aids, devices, their configurations, and the possible combinations between disabilities and configurations were analyzed within a custom waterfall ontology engineering methodology, in which the expertise of INAIL personnel was fundamental to disentangle the implications of the laws and, thus, clarifying the domain. The engineering process is composed of three main steps: domain analysis and conceptualization, prototype development, and testing.
Due to its “persistent” aspect, the knowledge related to the PwD’s health condition and the legal vehicle modifications are not subject to frequent changes; therefore, the DSS—and its underlying ontology—do not need to manage real-time data. On the contrary, the knowledge related to a PwD’s health is obtained by consulting clinical personnel, who are the only stakeholders in charge of compiling the PwD’s health condition profile (adopting the formalisms described in the following paragraphs).
4.1.1. Domain Analysis and Conceptualization
The activities conducted in this step are paramount in any ontology engineering process since they enable the discovery of the relevant knowledge and its implications and provide it with a conceptual form [36,38]. Since, as detailed in the previous section, the European and Italian regulations discipline the domain of vehicular modifications for PwDs, the efforts were related to the identification of the norms, their structure, and the combinatory rules. The regulations considered explicitly indicate that a driver can be affected by a single impairment or a double impairment, i.e., a disability affecting two limbs (whether they are both superior or lower limbs). The total number of single impairment conditions foreseen by the Italian norm is 14, while the number of double impairment conditions amounts to 130. The impairment codes are named with an integer number from 1 to 5, with numbers from 1 to 3 devoted to the description of upper limb conditions and numbers 4 and 5 dedicated to lower limb conditions. Each integer can then be further specified with a capital letter (or a couple of capital letters) to specify the impairment and its severity further. Table 3 provides an example of the use of integer numbers combined with letters to describe the conditions.
Table 3 illustrates the conditions related to the impairment affecting one arm, with letters A, B, and C further characterizing the type and residual functionalities resulting from the disability. In addition to the above-mentioned capital letters, codes 2 and 3 (devoted to arm and hand impairments descriptions) can also be characterized using the PE capital letters (standing for “aesthetic prosthesis”). Codes 4 and 5 describe, respectively, the loss or complete lack of functionality of a lower limb and a partial anatomical loss or partial lack of functionality of a lower limb; these codes do not require to be characterized with capital letters (except for 5SF, indicating that the loss of functionality of the lower limb does not compromise the possibility to use brake and accelerator pedals correctly).
The set of double impairment codes combines all the possibilities expressed by the single impairment codes in couples for a total of 130 double impairment codes. Thus, these codes can be used to describe impairments characterizing both arms, both legs, or one arm and one leg and to specify the severity of the anatomical or functional loss for each limb.
Regarding the aids and devices that could be adopted in car configurations, according to the norms (see Table 2) it is possible to divide them into three broad categories: aids for the drivers, vehicle adaptations, and restrictions on vehicle use. The first category encompasses the codes from 01 to 05, which correspond to medically prescribed devices (such as glasses, hearing aids, prostheses, etc.); the second category includes all the codes related to the devices, aids, and modifications that could potentially be adopted in car configurations for PwDs (codes from 10 to 51 and all the included subcodes). Finally, the third category includes the codes from 61 to 69, which detail specific limitations to the use of the vehicle (prescribed by a medical committee). In this sense, the European norm is rigid and is a source of knowledge that is not open to any particular interpretation.
Similarly, the configurations are also rigidly normed by the 148/1991 Ministry of Transport circular. The vehicular configurations can be divided into left-side and right-side configurations, since (as highlighted in Figure 2 and Figure 3), the configurations vary considerably depending on the side where the impairment is located.
Taking into account all the above-mentioned considerations, the ontology must be able to provide an answer to the following Competency Questions (CQs) [39]:
CQ1: What are the impaired conditions foreseen by the norms for which a car configuration can be suggested?
CQ2: What configurations can be suggested for a PwD, according to his/her health condition?
CQ3: What devices or aids compose a car configuration?
CQ4: On which side of the car should the devices composing the configuration be located?
4.1.2. Development
Following the domain analysis and conceptualization phase, the conceptualization acquired enabled the development of the ontology. To model the knowledge base, the W3C-endorsed languages Resource Description Framework (RDF) [40] and Ontology Web Language (OWL) [41], together with the Semantic Web Rule Language (SWRL) [42] and the query language SPARQL Protocol and RDF Query Language (SPARQL) [43] were used.
These languages are adopted in a variety of ontology-based DSSs and enable ontological reasoning [26], which resembles human inferential abilities. In detail, the OWL profile selected for the Rip@rto ontology is OWL 2 DL, which provides a fair tradeoff between knowledge representation and reasoning capabilities [44]. The possibility of exploiting ontological reasoning capabilities is pivotal in health-related research fields (as pointed out in current literature reviews [45] and in recent works [32,33,34,46]).
The structure of the ontology (prefixed with
As highlighted in Figure 5, the class
Similarly, the class
The
For each subclass belonging to
As illustrated above and in Figure 6, each class was instantiated, including the
Finally, the development is completed with a set of SWRL rules, which allows for an indication on which condition(s) a car configuration should be suggested to a PwD. For example, the rule:
[1]
Candidate_driver(?d) ∧hasImpairmentCondition(?d, ?hc) ∧1Bs + 5d(?hc) →suggestedSetUp(?d, conf_45B) ∧suggestedSetUp(?d, conf_46)
Similarly, the rule:
[2]
Candidate_driver(?d) ∧hasImpairmentCondition(?d, ?hc) ∧4s + 5d(?hc) →suggestedSetUp(?d, conf_57A) ∧suggestedSetUp(?d, conf_57B) ∧suggestedSetUp(?d, conf_57C)
The development of SWRL rules relies on domain experts and legislative decrees’ knowledge: the matching between specific health conditions and the possible configurations a vehicle can assume is detailed in updating the 148/1991 Ministry of Transport circular. Thus, the Rip@rto ontology “translates” the portions of the law into SWRL rules; the process is eased by the tables depicted within the Italian circular, which are very similar to those illustrated in Figure 2 and Figure 3. The role of domain experts and the cooperation with ontology engineers is pivotal to support the correct interpretation of the regulations and to enable correct SWRL rules definition [37,38,47].
Moreover, rules and
More than 165 rules are developed to cover all matchings between a PWD’s health condition (represented by the
The ontology is published online (accessible on:
4.1.3. Test
The Rip@rto ontology described above is then populated with instances representing some PwDs and their
The aim of developing individuals was the test of the ontological prototype (and its set of rules). Also, the test needs to verify whether the developed Rip@rto ontology can answer the CQs presented in the previous subsection. The four users are characterized as follows:
TestUser_01: impairment condition 1Bs + 5d (left impaired arm capable of gripping the steering wheel and operating controls and a partial loss of functionality of the right leg)
TestUser_02: impairment condition 4s + 5d (a complete loss of functionality on the left leg and a partial loss of functionality on the right leg)
TestUser_03: impairment condition 1PM (loss of an arm, specified to be on the right side and equipped with myoelectric prosthesis)
TestUser_04: impairment condition 1Cd + 4d (functional limitations to the right arm with enough abilities for driving and a total loss of functionality on the right leg).
The CQs presented above were formulated with SPARQL as follows (Table 4):
The CQs 2 and 3 could be conveniently grouped into one SPARQL query (although CQ4 could have also been grouped in the same query, it was presented separately to underline that the information pertaining to the side on which aids are located is independent from a PwD’s health condition).
The SPARQL queries illustrated allow us to show the configurations inferred for each of the four test users, the aids and their location, and the candidate drivers with disabilities with the impairment(s) characterizing them. The inferences drawn were considered correct and adequate by INAIL personnel. Similarly to other existing knowledge-based DSSs in the healthcare field [33,34,49], we decided to ask experts to check, validate, and integrate the results of the inference process. No modifications to the inferred information were required. Therefore, the test phase is completed, and the Rip@rto ontology is published.
4.2. Rip@rto Application
In order to help drivers with disabilities better visualize the configurations and required car modifications, a simple application is designed and implemented as a GUI. The application is developed with Unity 3D engine and C# scripting language for cross-platform support. The application not only helps with visualization but also serves as a middleware between the GUI and the semantic repository to dynamically communicate and exchange data. The application needs to access the ontology where the drivers’ health conditions and disabilities are defined to receive the inference results about each driver’s health condition, disabilities, and limitations. The reasoned and inferred results about the explicit list of car configurations and driving aids for each driver can then be visualized within the application. Additionally, the application must be capable of modifying the ontology in case a driver’s health condition has changed or a new driver must be inserted into the ontology.
However, modifying the ontology on the server is not a trivial task because of the Open-World Assumption (OWA) of the monotonic nature of Description Logic (DL) [50]. Reasoning in the OWA means that the statements about knowledge that are not included in or inferred from the knowledge explicitly modeled in the ontology may be considered unknown rather than wrong or false. This makes it impossible for the deductive reasoner to infer the existence of a new instance unless it is already modeled in the knowledge base. Therefore, the application needs to be able to dynamically connect to the semantic database where the ontology is stored, retrieve data from the ontology, and insert new pieces of information into the ontology when needed.
The domain ontologies are stored on the Stardog server, which is a semantic repository that enables semantic reasoning with OWL, capable of managing rules in the SWRL format, and it can export inferred data in various formats.
The application generates a proper SPARQL query ‘INSERT’ and runs the query on the Stardog reasoner to upload the ontology on the Stardog semantic repository. It can generate the same query later again in case the ontology needs to be updated with a new driver or modified information. It then proceeds to run the reasoner to materialize the most recent inferences based on the SWRL rules that have been defined and stored on the Stardog server with the ontology.
When the ontology is successfully uploaded and reasoned on the Stardog repository, the application is ready to be used by the users (with the support of INAIL personnel). The application displays a list of candidate drivers fetched from the ontology so the user can choose the driver they want to analyze. After choosing the driver on the application, the next page is going to list all the configurations that would work properly for the candidate driver chosen (Figure 7). This list is generated as a result of the application generating and running the proper SPARQL query ‘SELECT’ in the background to fetch the configuration data regarding the chosen candidate driver. Once the user chooses the configuration, the application displays all the necessary driving aids and car modifications together with an image of the installed aids on the car to better visualize how their car will look with the final result (Figure 8).
5. Discussion
The proposed DSS is developed with the aim of supporting drivers with disabilities in identifying the possible aids, modifications, and devices that could characterize their vehicles. Such devices and modifications are thought to support PwD in driving by helping them in coping with their impairments while leveraging on their residual abilities.
The system is developed starting from the existing regulatory framework, which indicates the possible disabilities (and their combinations) and classifies the available modifications, devices, and aids. Since this knowledge is strictly regulated, the DSS adopted ontology as a way to formalize the necessary information; its fruition is exerted with an application, here presented as a prototype. This prototypical application helps the users communicate with the semantic repository and exchange information in real time without having prior knowledge of ontologies. It provides a simple GUI in which all the reasoned and inferred data from the ontology can be translated into human-readable information together with graphical representations of the vehicle for better visual understanding.
Regarding the Rip@rto knowledge base, it is worth noting some interesting effects pertaining to the choice of adopting an ontology. First, the semantic representation is suited to represent classifications, particularly those characterized by elements and sub-elements (the characterization of the European and Italian norms). The devices, aids, and mechanical modifications foreseen by the EU Directive 2015/653 enable the classification of any device that could be employed to modify a PwD’s vehicle. In this way, this portion of the ontology reflects the regulatory framework: each new device would be necessarily classified according to the existing classification; hence, the classes represented in the ontology will be more than adequate for this purpose. A similar consideration can be drawn for the classification of impairments. The set of the single and double impairments foreseen by the regulatory framework is unlikely to change, unless the basic physical requirements necessary for drivers with disabilities to drive safely are discussed again. Therefore, the description of the impairments the ontology provides is exhaustive and can be easily adapted for any European context.
However, the Rip@rto ontology—although pretty static—is far from immune from the necessity of being updated. For example, if we consider the development of new devices with different functionalities from the existing ones, they would require the amendment of the classification of the EU Directive 2015/653; thus, the Rip@rto ontology would also require an update. As a consequence, the set of configurations could also vary since new devices may combine the functionalities of one or more existing ones. Nonetheless, the Rip@rto ontology can be updated and tested (limited to the modifications generated).
A relevant aspect in which the DSS could support the clinical personnel is the identification of restrictions on use codes. The Rip@rto DSS does not indicate any of these codes, which are not devices, so they do not fall under the scope of the DSS. However, it is not infrequent that these codes are used in daily life to limit the use of cars for those individuals whose abilities are somehow limited. In this regard, the ontological layer could take advantage of the results of driving simulators for PwDs (in particular, the one that is being developed in the Rip@rto project [17]) to suggest clinicians with a set of “restriction on use” codes. This can potentially be achieved by acquiring and processing simulations’ data related to assessment and training and by adding to the ontology a second layer of SWRL rules.
6. Conclusions
This work introduced the Rip@rto DSS, a prototypical ontology-based application to support PwDs in identifying and becoming accustomed to the aids, mechanical modifications, and devices that could characterize their vehicles. The ontology underlying the DSS is based on European (and Italian) norms and was developed in close collaboration with the Italian National Institute for Insurance against Accidents at Work (INAIL). By relying on ontological reasoning, the DSS is capable of providing drivers with disabilities with the possible sets of modifications their vehicles could mount and, thus, enabling them to make a conscious choice among the different possibilities. For each possible configuration, the DSS also illustrates the side of the vehicle where the aids are going to be placed.
A prototypical application demonstrates the potential of this approach, indicating different possible car configurations for each driver and illustrating within the vehicle cabin where the devices are going to be deployed for each configuration.
This work falls under the category of systems devoted to supporting PwDs in re-acquiring their ability to drive and, thus, regaining their autonomy in life. Nonetheless, it is necessary to mention that while a DSS such as Rip@rto may support PwDs and clinicians in the short and medium term, the rapid development of assisted driving systems may serve as a long-term solution (not limited to PwDs) [51,52,53].
Future works on Rip@rto foresee the possibility of further expanding the ontological knowledge base to support the identification of restrictions on the use codes. In this sense, the ontological layer could serve as part of a more complex (and, possibly, automated [54]) system devoted to assessing candidate drivers and suggesting vehicular modifications. Nevertheless, before modifying the application, it is necessary to assess Rip@rto’s acceptance among clinicians composing the medical committees and candidate drivers. To this aim, a set of validation tests will be set in place, exploiting the Technology Acceptance Model (TAM3) [55] and interviews to gather feedback from the end users of the DSS. Also, the application will be improved to match the expanded ontology and provide faster data exchange and better visualization.
Finally, the prototype will be tested with both candidate drivers (PwDs) and professionals from the medical committees. The aims of the tests will be: (a) to acquire the usability scores of the prototype and (b) to acquire feedback to further enhance both the ontological layer and the interface. The tests will involve the administration of the System Usability Scale (SUS) [56] and the Technology Acceptance Model [55] to assess the DSS’s general usability an to further investigates those aspects pertaining to the possible adoption of such a system in the real world by clinicians.
Conceptualization, D.S. and A.M.; methodology, D.S. and A.M.; software, D.S. and A.M.; investigation, D.S.; resources, M.S. and A.D.; writing—original draft preparation, D.S. and A.M.; writing—review and editing, M.S.; visualization, D.S. and A.M.; supervision, D.S.; project administration, M.S. and A.D.; funding acquisition, M.S. and A.D. All authors have read and agreed to the published version of the manuscript.
The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author/s.
The authors declare no conflicts of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 2. An example of the impairments (a single impairment and a double impairment) and the possible configurations foreseen by the Italian law.
Enlarge this image.Figure 3. An example of optional vehicle modifications depending on the type of disability and its location within the body.
Enlarge this image.Figure 4. Conceptual architecture of the Rip@rto DSS, including the application and ontology.
Enlarge this image.Figure 6. A portion of the ABox and TBox of the Rip@rto ontology; in the example, the class r:15.01 (Adapted clutch pedal) and its corresponding individual are shown: the individual is linked to the configurations’ individuals in which it is foreseen.
Enlarge this image.Figure 7. A screenshot of the application illustrating the three possible car configurations (conf_57A, conf_57B, and conf_57C) displaying all the aids necessary for each configuration for the candidate driver “Tazio”.
Enlarge this image.Figure 8. A screenshot of the application visualizing the position of the car modifications and driving aids for the chosen configuration (a); each device available can be further detailed to illustrate its description (b).
The list of included works with the illustration of their main topics.
| Year | Type | Topic | Description | |
|---|---|---|---|---|
| [ | 1989 | Conference | DSS | Clinical DSS for the identification of equipment |
| [ | 2021 | Journal article | Equipment | Study of a feature for equipment |
| [ | 2022 | Journal article | Simulator | Training and assessment |
| [ | 2022 | Conference | DSS | PwD-dedicated DSS for the selection of equipment |
| [ | 2023 | Conference | Equipment | Equipment to support coping |
| [ | 2023 | Journal article | Equipment | Design of a piece of equipment |
| [ | 2023 | Conference | Simulator | Training and assessment |
| [ | 2023 | Conference | DSS | PwD-dedicated DSS for the selection of equipment |
| [ | 2024 | Conference | Simulator | Training and assessment |
An excerpt of the codes foreseen by the EU Directive 2015/653.
| Specific | Code | Name |
|---|---|---|
| Driver (medical reason) | 01 | Sight correction and/or protection |
| 02 | Hearing aid/communication aid | |
| 03 | Prosthesis/orthosis for the limbs | |
| 05 | Limited use (subcode use obligatory, driving subject to restrictions for medical reasons) | |
| Vehicle adaptations | 10 | Modified transmission |
| 15 | Modified clutch | |
| 20 | Modified braking system | |
| 25 | Modified accelerator system | |
| 31 | Pedal adaptations and pedal safeguards | |
| 32 | Combined service brake and accelerator systems | |
| 33 | Combined service brake, accelerator and steering systems | |
| 35 | Modified control layouts (Lights switches, windscreen wiper/washer, horn, direction indicators, etc.) | |
| 40 | Modified steering | |
| 42 | Modified rearview mirror(s) | |
| 43 | Driver seating position | |
| 44 | Modifications to motorcycles (subcode use obligatory) | |
| 45 | Motorcycle with sidecar only | |
| 46 | Tricycles only | |
| 47 | Restricted to vehicles of more than two wheels not requiring balance by the driver for starting, stopping and standing | |
| 50 | Restricted to a specific vehicle/chassis number (vehicle identification number, VIN) | |
| 51 | Restricted to a specific vehicle/chassis number (vehicle identification number, VIN) | |
| Limited use codes | 61 | Limited to daytime journeys (for example: one hour after sunrise and one hour before sunset) |
| 62 | Limited to journeys within a radius of…km from holder’s place of residence or only inside city/region | |
| 63 | Driving without passengers | |
| 64 | Limited to journeys with a speed not greater than…km/h | |
| 65 | Driving authorized solely when accompanied by a holder of a driving license of at least the equivalent category | |
| 66 | Without trailer | |
| 67 | No driving on motorways | |
| 68 | No alcohol | |
| 69 | Restricted to driving vehicles equipped with an alcohol interlock in accordance with EN 50436 |
Examples of the impairment codes, their names, and their description.
| Code | Name | Description |
|---|---|---|
| 1 | No functionality in the mutilated or impaired limb | Total loss or agenesis of one of the upper limbs OR functional limitation of one of the upper limbs that does not allow the use of any driving control. |
| 1A | Impaired limb capable of grip on the steering wheel only | Functional limitation to one of the upper limbs (due to stabilized outcomes of an injury) which nevertheless allows to grip the steering wheel with autonomous movement and secure grip, but without the possibility of using the gearshift (manual transmission) and without the possibility to operate controls on the side of the impaired limb. |
| 1B | Impaired limb capable of grip on the steering wheel and to operate controls | Functional limitation to one of the upper limbs (due to stabilized outcomes of an injury) which nevertheless allows to grip the steering wheel with autonomous movement and secure grip AND to operate controls on the side of the impairment, but without the possibility of using the gearshift (manual transmission). |
| 1C | Impaired limb but capable of all functions required for driving | Functional limitation to one of the upper limbs (due to stabilized outcomes of an injury) which nevertheless allows to grip the steering wheel with autonomous movement and secure grip, AND to operate controls on the side of the impairment, AND to use the gearshift (manual transmission, if the impairment is on the right), AND to use the parking brake lever (if the impairment is on the right). |
| 1PM | Limb with myoelectric prosthesis | Anatomical loss or agenesis of one of the upper limbs at a higher level of the elbow joint and with conservation of the scapulohumeral joint, compensated with a tolerated and efficient myoelectric prosthesis that allows fine hand and elbow’s movements necessary for driving. |
The SPARQL queries “translating” the CQ.
| CQ | Question | SPARQL |
|---|---|---|
| CQ1 | What are the impaired conditions foreseen by the norms for which a car configuration can be suggested? | [Image omitted. Please see PDF.] |
| CQ2, CQ3 | What configurations can be suggested for a PwD, according to his/her health condition | [Image omitted. Please see PDF.] |
| CQ4 | On which side of the car should the devices composing the configuration be located? | [Image omitted. Please see PDF.] |
References
1. Hutchinson, C.; Berndt, A.; Gilbert-Hunt, S.; George, S.; Ratcliffe, J. Modified Motor Vehicles: The Experiences of Drivers with Disabilities. Disabil. Rehabil.; 2020; 42, pp. 3043-3051. [DOI: https://dx.doi.org/10.1080/09638288.2019.1583778] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30907156]
2. Trujillo, L.G. Driver Rehabilitation and Community Mobility Principles and Practice. Occup. Ther. Health Care; 2008; 22, pp. 93-94. [DOI: https://dx.doi.org/10.1080/07380570802244621] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23930768]
3. Dahuri, M.K.A.M.; Hussain, M.N.; Yusof, N.F.M.; Jalil, M.K.A. Factors, Effects, and Preferences on Vehicle Driving Modification for the Malaysia Independent Disabled Driver. J. Soc. Automot. Eng. Malays.; 2017; 1, pp. 103-110. [DOI: https://dx.doi.org/10.56381/jsaem.v1i2.13]
4. Darcy, S.; Burke, P.F. On the Road Again: The Barriers and Benefits of Automobility for People with Disability. Transp. Res. Part. A Policy Pract.; 2018; 107, pp. 229-245. [DOI: https://dx.doi.org/10.1016/j.tra.2017.11.002]
5. Di Stefano, M.; Stuckey, R.; Kinsman, N. Understanding Characteristics and Experiences of Drivers Using Vehicle Modifications. Am. J. Occup. Ther.; 2019; 73, pp. 7301205050p1-7301205050p9. [DOI: https://dx.doi.org/10.5014/ajot.2019.023721]
6. Monacelli, E.; Dupin, F.; Dumas, C.; Wagstaff, P. A Review of the Current Situation and Some Future Developments to Aid Disabled and Senior Drivers in France. IRBM; 2009; 30, pp. 234-239. [DOI: https://dx.doi.org/10.1016/j.irbm.2009.09.004]
7. Murray-Leslie, C. Aids for Disabled Drivers. BMJ; 1990; 301, pp. 1206-1209. [DOI: https://dx.doi.org/10.1136/bmj.301.6762.1206]
8. Di Stefano, M.; Stuckey, R.; Macdonald, W.; Lavender, K. Vehicle Modifications for Drivers with Disabilities: Developing the Evidence Base to Support Prescription Guidelines, Improve User Safety and Enhance Participation; La Trobe University–Melbourne: Melbourne, Australia, 2015.
9. Ysander, L. The Safety of Physically Disabled Drivers. Occup. Environ. Med.; 1966; 23, pp. 173-180. [DOI: https://dx.doi.org/10.1136/oem.23.3.173]
10. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E. et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ; 2021; n71. [DOI: https://dx.doi.org/10.1136/bmj.n71]
11. Wheeler, J.C.; Koppa, R.; McDermott, M.; Ellis, N.; Huchingson, R.D. A Knowledge Based System to Suggest Adaptive Equipment for Disabled Drivers. Proceedings of the Images of the Twenty-First Century. Proceedings of the Annual International Engineering in Medicine and Biology Society; Seattle, WA, USA, 9–12 November 1989; pp. 1861-1862.
12. Dols, J.F.; Girbes-Juan, V.; Jimenez, I. On the Assessment of Fitness to Drive: Steering and Brake Operative Forces. IEEE Access; 2021; 9, pp. 134682-134694. [DOI: https://dx.doi.org/10.1109/ACCESS.2021.3116080]
13. Ricci, S.; Gandolfi, F.; Marchesi, G.; Bellitto, A.; Basteris, A.; Canessa, A.; Massone, A.; Casadio, M. ADRIS: The New Open-Source Accessible Driving Simulator for Training and Evaluation of Driving Abilities. Comput. Methods Programs Biomed.; 2022; 221, 106857. [DOI: https://dx.doi.org/10.1016/j.cmpb.2022.106857]
14. Spoladore, D.; Cilsal, T.; Mahroo, A.; Trombetta, A.; Sacco, M. Towards an Ontology-Based Decision Support System to Support Car-Reconfiguration for Novice Wheelchair Users. International Conference on Computers Helping People with Special Needs; Springer International Publishing: Cham, Switzerland, 2022; pp. 445-452.
15. Salem, O.; Mehaoua, A.; Boutaba, R. The Sight for Hearing: An IoT-Based System to Assist Drivers with Hearing Disability. Proceedings of the 2023 IEEE Symposium on Computers and Communications (ISCC); Gammarth, Tunisia, 9–12 July 2023; pp. 1305-1310.
16. Suliano, S.B.; Ahmad, S.A.; As’arry, A.; Abdul Aziz, F. Conceptual Design of a Combined Brake-Accelerator Pedal for Limbs Disabled Driver Using a Hybrid Approach. Pertanika J. Sci. Technol.; 2023; 31, pp. 895-909. [DOI: https://dx.doi.org/10.47836/pjst.31.2.12]
17. Arlati, S.; Colombo, V.; Davalli, A.; Improta, M.; Ortolani, N.; Sacco, M. A Scalable VR-Based System to Assess Driving-Related Abilities in People with Disability. Proceedings of the 2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE); Milano, Italy, 25–27 October 2023; pp. 92-97.
18. Mahroo, A.; Spoladore, D.; Davalli, A. An Ontology-Based Mixed Reality Application to Support Car Reconfiguration for Drivers with Disabilities. Proceedings of the 2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE); Milano, Italy, 25–27 October 2023; pp. 6-10.
19. Adam Jones, J.; Woody Watson, W.; Stewart, T.; Brewington, J.; Chrismond, C.; Dabbiru, L.; Ahonle, Z.J.; Wall, E.S.; Geroux, K.; Stratton-Gadke, K. DriVR: Extending Driver Training for Persons with Disabilities. Proceedings of the 2024 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW); Orlando, FL, USA, 16–21 March 2024; pp. 1218-1219.
20. European Commission Commission Directive (EU) 2015/653 of 24 April 2015 Amending Directive 2006/126/EC of the European Parliament and of the Council on Driving Licences. Available online: https://Eur-Lex.Europa.Eu/Eli/Dir/2015/653/Oj (accessed on 17 October 2024).
21. Chen, Z.; Zhan, G.; Zhifan, J.; Zhang, W.; Rao, Z.; Wang, H.; Li, J. Adaptive Impedance Control for Docking Robot via Stewart Parallel Mechanism. ISA Trans.; 2024.in press [DOI: https://dx.doi.org/10.1016/j.isatra.2024.09.008] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39368867]
22. Xu, Y.; Chen, Z.; Deng, C.; Wang, S.; Wang, J. LCDL: Toward Dynamic Localization for Autonomous Landing of Unmanned Aerial Vehicle Based on LiDAR–Camera Fusion. IEEE Sens. J.; 2024; 24, pp. 26407-26415. [DOI: https://dx.doi.org/10.1109/JSEN.2024.3424218]
23. Italian Government Italian Government: Decree 285/1992 (and Successive Modifications) “New Highway Code, Road Safety and Vehicle Use in Italy”. Available online: https://Www.Normattiva.It/Uri-Res/N2Ls?Urn:Nir:Stato:Decreto.Legislativo:1992-04-30;285!Vig= (accessed on 17 October 2024).
24. Meskó, B.; Drobni, Z.; Bényei, É.; Gergely, B.; Győrffy, Z. Digital Health Is a Cultural Transformation of Traditional Healthcare. Mhealth; 2017; 3, 38. [DOI: https://dx.doi.org/10.21037/mhealth.2017.08.07]
25. Musacchio, N.; Giancaterini, A.; Guaita, G.; Ozzello, A.; Pellegrini, M.A.; Ponzani, P.; Russo, G.T.; Zilich, R.; de Micheli, A. Artificial Intelligence and Big Data in Diabetes Care: A Position Statement of the Italian Association of Medical Diabetologists. J. Med. Internet Res.; 2020; 22, e16922. [DOI: https://dx.doi.org/10.2196/16922]
26. Bienvenu, M.; Leclère, M.; Mugnier, M.-L.; Rousset, M.-C. Reasoning with Ontologies. A Guided Tour of Artificial Intelligence Research; Springer International Publishing: Cham, Switzerland, 2020; pp. 185-215.
27. Earley, S. The Problem with AI. IT Prof.; 2017; 19, pp. 63-67. [DOI: https://dx.doi.org/10.1109/MITP.2017.3051331]
28. Dzemydiene, D.; Kazemikaitiene, E. Ontology-Based Decision Support System for Crime Investigation Processes. Information Systems Development; Springer: New York, NY, USA, 2005; pp. 427-438.
29. Wulandari, I.A.; Sensuse, D.I.; Krisnadhi, A.A.; Akmaliah, I.F.; Rahayu, P. Ontologies for Decision Support System: The Study of Focus and Techniques. Proceedings of the 2018 10th International Conference on Information Technology and Electrical Engineering (ICITEE); Bali, Indonesia, 24–26 July 2018; pp. 609-614.
30. Tudorache, T. Ontology Engineering: Current State, Challenges, and Future Directions. Semant. Web; 2020; 11, pp. 125-138. [DOI: https://dx.doi.org/10.3233/SW-190382]
31. Jing, X.; Min, H.; Gong, Y.; Biondich, P.; Robinson, D.; Law, T.; Nohr, C.; Faxvaag, A.; Rennert, L.; Hubig, N. et al. Ontologies Applied in Clinical Decision Support System Rules: Systematic Review. JMIR Med. Inform.; 2023; 11, e43053. [DOI: https://dx.doi.org/10.2196/43053]
32. Spoladore, D.; Stella, F.; Tosi, M.; Lorenzini, E.C.; Bettini, C. A Knowledge-Based Decision Support System to Support Family Doctors in Personalizing Type-2 Diabetes Mellitus Medical Nutrition Therapy. Comput. Biol. Med.; 2024; 180, 109001. [DOI: https://dx.doi.org/10.1016/j.compbiomed.2024.109001]
33. Spoladore, D.; Colombo, V.; Fumagalli, A.; Tosi, M.; Lorenzini, E.C.; Sacco, M. An Ontology-Based Decision Support System for Tailored Clinical Nutrition Recommendations for Patients With Chronic Obstructive Pulmonary Disease: Development and Acceptability Study. JMIR Med. Inform.; 2024; 12, e50980. [DOI: https://dx.doi.org/10.2196/50980] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38922666]
34. Spoladore, D.; Negri, L.; Arlati, S.; Mahroo, A.; Fossati, M.; Biffi, E.; Davalli, A.; Trombetta, A.; Sacco, M. Towards a Knowledge-Based Decision Support System to Foster the Return to Work of Wheelchair Users. Comput. Struct. Biotechnol. J.; 2024; 24, pp. 374-392. [DOI: https://dx.doi.org/10.1016/j.csbj.2024.05.013] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38800691]
35. Gruber, T.R. A Translation Approach to Portable Ontology Specifications. Knowl. Acquis.; 1993; 5, pp. 199-220. [DOI: https://dx.doi.org/10.1006/knac.1993.1008]
36. Simperl, E.; Luczak-Rösch, M. Collaborative Ontology Engineering: A Survey. Knowl. Eng. Rev.; 2014; 29, pp. 101-131. [DOI: https://dx.doi.org/10.1017/S0269888913000192]
37. Spoladore, D.; Pessot, E. Collaborative Ontology Engineering Methodologies for the Development of Decision Support Systems: Case Studies in the Healthcare Domain. Electronics; 2021; 10, 1060. [DOI: https://dx.doi.org/10.3390/electronics10091060]
38. Spoladore, D.; Pessot, E. An Evaluation of Agile Ontology Engineering Methodologies for the Digital Transformation of Companies. Comput. Ind.; 2022; 140, 103690. [DOI: https://dx.doi.org/10.1016/j.compind.2022.103690]
39. Grüninger, M.; Fox, M.S. The Role of Competency Questions in Enterprise Engineering. Benchmarking—Theory and practice; Springer: Boston, MA, USA, 1995.
40. Pan, J.Z. Resource Description Framework. Handbook on Ontologies; Springer: Berlin/Heidelberg, Germany, 2009; pp. 71-90.
41. Antoniou, G.; Harmelen, F.V. Web Ontology Language: OWL. Handbook on Ontologies; Springer: Berlin/Heidelberg, Germany, 2009.
42. Horrocks, I.; Patel-Schneider, P.F.; Boley, H.; Tabet, S.; Grosof, B.; Dean, M. Others SWRL: A Semantic Web Rule Language Combining OWL and RuleML. W3C Memb. Submiss.; 2004; 21, pp. 1-31.
43. Hommeaux, E.P.; Seaborne, A. SPARQL Query Language for RDF. W3C Recommendation. 2008; Available online: http://www.w3.org/TR/rdf-sparql-query/ (accessed on 17 October 2024).
44. Krötzsch, M. OWL 2 Profiles: An Introduction to Lightweight Ontology Languages. Reasoning Web International Summer School; Springer: Berlin/Heidelberg, Germany, 2012; pp. 112-183.
45. Dissanayake, P.I.; Colicchio, T.K.; Cimino, J.J. Using Clinical Reasoning Ontologies to Make Smarter Clinical Decision Support Systems: A Systematic Review and Data Synthesis. J. Am. Med. Inform. Assoc.; 2020; 27, pp. 159-174. [DOI: https://dx.doi.org/10.1093/jamia/ocz169]
46. Spoladore, D.; Stella, F.; Tosi, M.; Lorenzini, E.C. Towards a Knowledge-Based Decision Support System for the Management of Type 2 Diabetic Patients. International Symposium on Industrial Engineering and Automation; Springer Nature: Cham, Switzerland, 2023; pp. 309-320.
47. Spoladore, D.; Pessot, E.; Trombetta, A. A Novel Agile Ontology Engineering Methodology for Supporting Organizations in Collaborative Ontology Development. Comput. Ind.; 2023; 151, 103979. [DOI: https://dx.doi.org/10.1016/j.compind.2023.103979]
48. World Wide Web Consortium. RDF 1.1 Turtle: Terse RDF Triple Language. 2014; Available online: https://travesia.mcu.es/server/api/core/bitstreams/cb5a63fb-a59d-4951-81a3-55bba298353c/content (accessed on 17 October 2024).
49. Spoladore, D.; Colombo, V.; Campanella, V.; Lunetta, C.; Mondellini, M.; Mahroo, A.; Cerri, F.; Sacco, M. A Knowledge-Based Decision Support System for Recommending Safe Recipes to Individuals with Dysphagia. Comput. Biol. Med.; 2024; 171, 108193. [DOI: https://dx.doi.org/10.1016/j.compbiomed.2024.108193]
50. Baader, F.; Horrocks, I.; Sattler, U. Description Logics as Ontology Languages for the Semantic Web. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Springer: Berlin/Heidelberg, Germany, 2005; Volume 2605, [DOI: https://dx.doi.org/10.1007/978-3-540-32254-2_14]
51. Liang, J.; Lu, Y.; Yin, G.; Fang, Z.; Zhuang, W.; Ren, Y.; Xu, L.; Li, Y. A Distributed Integrated Control Architecture of AFS and DYC Based on MAS for Distributed Drive Electric Vehicles. IEEE Trans. Veh. Technol.; 2021; 70, pp. 5565-5577. [DOI: https://dx.doi.org/10.1109/TVT.2021.3076105]
52. Wang, Z.; Han, K.; Tiwari, P. Digital Twin-Assisted Cooperative Driving at Non-Signalized Intersections. IEEE Trans. Intell. Veh.; 2022; 7, pp. 198-209. [DOI: https://dx.doi.org/10.1109/TIV.2021.3100465]
53. Liang, J.; Lu, Y.; Wang, F.; Yin, G.; Zhu, X.; Li, Y. A Robust Dynamic Game-Based Control Framework for Integrated Torque Vectoring and Active Front-Wheel Steering System. IEEE Trans. Intell. Transp. Syst.; 2023; 24, pp. 7328-7341. [DOI: https://dx.doi.org/10.1109/TITS.2023.3262655]
54. Dicianno, B.E.; Sivakanthan, S.; Sundaram, S.A.; Satpute, S.; Kulich, H.; Powers, E.; Deepak, N.; Russell, R.; Cooper, R.; Cooper, R.A. Systematic Review: Automated Vehicles and Services for People with Disabilities. Neurosci. Lett.; 2021; 761, 136103. [DOI: https://dx.doi.org/10.1016/j.neulet.2021.136103]
55. Venkatesh, V.; Bala, H. Technology Acceptance Model 3 and a Research Agenda on Interventions. Decis. Sci.; 2008; 39, pp. 273-315. [DOI: https://dx.doi.org/10.1111/j.1540-5915.2008.00192.x]
56. Grier, R.A.; Bangor, A.; Kortum, P.; Peres, S.C. The System Usability Scale. Proc. Human. Factors Ergon. Soc. Annu. Meet.; 2013; 57, pp. 187-191. [DOI: https://dx.doi.org/10.1177/1541931213571042]
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Atieh Mahroo 2
; Davalli, Angelo 3 ; Sacco, Marco 1 1 National Research Council of Italy, Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing, 23900 Lecco, Italy;
2 National Research Council of Italy, Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing, 23900 Lecco, Italy;
3 National Institute for Insurance against Accidents at Work, 40054 Budrio, Italy