Robotic Assistance Research Articles

Improving the learning skills of autistic students using social robotics technologies

Teachers are instrumental in advancing the cognitive and motor skills of children with autism. Despite their importance, the incorporation of both educators and robotic aids in the educational frameworks of specialized schools and centers is infrequent. Extensive research has been conducted to evaluate the impact of robotic assistance on the learning outcomes for children with autism. This study investigates the effects of the Furhat robot on the educational experiences of autistic children in schools, analyzing its utility both with and without the presence of teachers. Interviews with educators were carried out to gauge the effectiveness of implementing Furhat robots in these settings. Data collected from sessions with autistic children were analyzed using ANOVA tests, offering insights into the Furhat Social Robot’s potential as a significant tool for fostering engagement and interaction. The findings highlight the robot’s effectiveness in enhancing social interaction and engagement, thereby contributing to the ongoing discussion on how social robots can improve the developmental progress and well-being of children with autism. Moreover, this paper underlines the innovative aspects of our proposed model and its wider implications. By presenting specific quantitative outcomes, our aim is to extend the reach of our findings to a broader audience. Ultimately, this research delineates significant contributions to the understanding of social robots, such as Furhat, in improving the overall well-being and developmental trajectories of children with autism.

Early case series with placement of NeuroOne Evo stereoelectroencephalography depth electrodes and review of other Food and Drug Administration-approved products

Background: Stereoelectroencephalography (SEEG) is a common diagnostic surgical procedure for patients with medically refractory epilepsy. We aimed to describe our initial experience with the recently released NeuroOne Evo SEEG electrode product (Zimmer Biomet, Warsaw, IN) and review technical specifications for other currently approved depth SEEG electrodes. Methods: We performed a record review on the first five patients implanted with NeuroOne Evo SEEG electrode product using the robotic stereotactic assistance robot platform and described our surgical technique in detail. We recorded technical specifications of all currently Food and Drug Administration-approved SEEG electrodes for comparison. Results: Our initial 5 surgical patients were reviewed. The average total time of operation was 92 min, with an average of 16.8 electrodes. The estimated time per electrode insertion was <2 min. There were no intracranial hemorrhages or hardware complications noted during monitoring. Monitoring provided diagnostic information in all patients, and removal and incision healing proceeded without issues. Conclusion: NeuroOne SEEG electrodes can be implanted with efficiency and provide a valuable additional tool for the epilepsy surgeon. A tapered drill bit prevents the bolt from being placed beyond the inner cortex and may reduce the risk of brain contusion or inadvertent advancement of anchor bolts, and the electrode internal stylet also affords the potential to reduce the number of trajectory passes. MeSH Terms: Epilepsy, EEG, Drug-resistant Epilepsy, Intracranial EEG

Exploring Robotic Nursing: A Comprehensive Systematic Review of Socially Assistive Robots within the Healthcare Professions

ABSTRACT Purpose The objective is to evaluate empirical literature on Socially Assistive Robots (SARs) in nursing, identifying strengths, weaknesses, and avenues for future research and practice. This article explores studies investigating robotic applications in healthcare, specifically nursing. Design The systematic review using PRISMA guidelines includes 19 relevant studies from Medline, PubMed, and Cochrane, presenting insights into the potential of SARs in addressing patients’ mental health and well-being. Methods The eligibility criteria for this review included publications from 2015 to 2024, full-text articles employing either qualitative or quantitative methodologies, articles written in English, and those published in peer-reviewed academic journals. Excluded were single-case studies, unpublished theses or dissertations, and studies not published in peer-reviewed journals. The search employed the terms “robotic nursing” and “future,” utilizing Medical Subject Headings (MeSH) terminology for precise retrieval. Findings The current literature has limitations in generalizability, breadth, and quantification of health-related outcomes. Areas of exploration, including diverse robot types, therapies, and geographical contexts, offer promising directions for future research. Conclusions The article concludes that although SARs show promise in healthcare, particularly in nursing, more comprehensive research is needed to establish their efficacy across a wider range of applications and settings. Clinical Evidence Current studies indicate that SARs may play a valuable role in supporting patients’ mental health and well-being, though further investigation is required to determine their broader impact and effectiveness.

Comparing the accuracy and treatment time of a robotic and dynamic navigation system in osteotomy and root-end resection: An invitro study.

To compare the accuracy and treatment time of a robotic system (RS) and a dynamic navigation system (DNS) in osteotomy and root-end resection in an invitro study. In this study, four identical mandibular models were created using three-dimensional (3D) printing, each model including 18 teeth. In the surgical procedures, teeth #35, #33, #32, #42, #43 and #45, along with the mesiobuccal roots of teeth #36 and #46, were specifically selected for operation on each model. Cone-beam computed tomography (CBCT) and intraoral scans were performed on all models. DentalNavi software was used to plan a drilling path and depth as well as design surgical accessories. In the RS, osteotomy and root-end resection were performed with robotic assistance, whereas in the DNS, these procedures were performed using dynamic navigation. Post-operative CBCT scans were obtained. Global platform deviation, global apex deviation, angular deflection, and time were compared to evaluate the accuracy and treatment time of the surgeries. The treatment time was defined as the registration time and drilling time. Statistical analyses were performed using the Shapiro-Wilk and independent sample t-test. The global platform deviation, global apex deviation and angular deflection in the RS group were significantly smaller than those in the DNS group (p <.05). The registration time in the RS group was significantly longer than that in the DNS group (p <.001), whereas the drilling time was significantly shorter (p <.001). The overall surgical time was significantly longer in the RS group than in the DNS group (p <.001). Subgroup accuracy analysis revealed that in both the anterior and posterior tooth regions, the RS group exhibited significantly smaller apical and angular deviations than the DNS group (p <.001). Additionally, in both the anterior and posterior tooth subgroups, the total time for the DNS group was significantly shorter than that for the RS group in the anterior tooth area (p <.01). RS showed higher accuracy and shorter drilling time in osteotomy and root-end resection compared to DNS. The RS holds promise as a reliable technological advancement in endodontic microsurgery. However, future endeavours should prioritize the reduction in registration time.

A Context-based Design Toolkit (CoDeT) for Socially Assistive Robots (SARs): a methodological approach

ABSTRACT This paper presents the Context-based Design Toolkit (CoDeT) methodology – a practical toolkit designed to facilitate collaborative design and evaluation of Socially Assistive Robots (SARs). CoDeT elicits user and other stakeholder needs, perceptions, and preferences. The methodology for creating a CoDeT is three-phased. The first phase is the Contextual space in which the designer generates a use-case-specific toolkit. The second phase is the Investigation space, which refers to utilising the toolkit among different users and stakeholders. The third phase is the Design Space, which returns to the design team to analyse the outcomes and define design guidelines. Rather than an exhaustive research investigation, our work provides guidelines and illustrative examples for using CoDeT effectively. Following the CoDeT methodology enables designers to gain insights regarding stakeholders’ and users’ perceptions and expectations of robotic roles and their most suitable appearances.

Towards a Hierarchical User Requirement Structure for Upper Limb Assistive Robotics

This article presents the results of an interdisciplinary study encompassing data collected from 108 in-depth interviews and 15 participatory workshops aimed at exploring the requirements of a diverse range of stakeholders in relation to assistive augmentative technologies. The study involved 39 post-stroke patients with hand paresis, 39 individuals with spinal cord injuries affecting upper limb function, 38 relatives and 65 healthcare professionals. Through qualitative analysis of the accumulated data, this research identified the multifaceted needs, preferences and aspirations of the study participants. We propose a structured hierarchy for user requirements, inspired by Maslow's hierarchy of needs. At the foundation of this pyramid are the essential functional requirements, forming the bedrock for the subsequent layers. Moving hierarchically, we identified reliability and security requirements, followed by usability and user experience considerations, recognition requirements and, finally, personal growth and self-fulfilment criteria. The foundational functional requirements must be fulfilled before other aspects can take precedence, creating a systematic approach to address the diverse needs of individuals with upper limb disabilities. The pyramid of requirements identified may significantly contribute to the scholarly community and lay the foundation for the development of assistive robotics that align with the desires of individuals with upper limb disabilities.

Why Would Professionals Choose to Use a Robot with Their Clients with Autism? Applying the Theory of Planned Behaviour to Determine Professionals’ Intentions

Robots are thought to be able to address some of society’s challenges in autism care and education. Robots may extend services and provide them in personalised, repeated, and playful ways. However, it is still largely unknown to what extent professionals intend to use robots and what determines their intentions. Using the Theory of Planned Behaviour framework, we aimed to better understand the determinants behind professionals’ intentions to use robots with their clients with autism. We conducted an online survey, to which 447 professionals (e.g., psychologists) working with people with autism answered. Our results indicate moderate and well-predicted intentions for the enquired professionals to use a robot with their clients with autism. We found that attitude and perceived social norms are the main predictors of their intentions. Our results also point to the significant and positive influence of moral norms on the attitudes of professionals. Additionally, our preliminary analysis of underlying beliefs to the assessed predictors of intention enables a better understanding on how professionals perceive the place of robots in their work. These results can inspire roboticists who want to build assistive robots for the real world and guide stakeholders when implementing such robots in the health and education sectors.

Ergonomic Considerations in Robotic-assisted Microsurgery

Objective: Microsurgery necessitates precision and skill, with conventional methods often leading to surgeon discomfort due to prolonged static postures. The introduction of robot-assisted microsurgery using the Symani surgical system and a three-dimensional exoscope aims to enhance surgical precision while mitigating these ergonomic issues. Methods: This prospective, single-center clinical study involved 50 preclinical training sessions and 87 robot-assisted microsurgical procedures at the authors' institution, covering a wide array of interventions from free flap reconstructions to nerve and lymphatic surgeries. Ergonomic assessment was conducted using the Localized Experienced Discomfort (LED) questionnaire, completed post-procedure. The study adhered to the Helsinki Declaration and received local ethics committee approval. Results: The neck and back region experienced the most discomfort (P &lt; 0.0001). Notably, operations performed with an exoscope showed significantly lower discomfort scores, particularly in the occipital and neck regions, compared with those using traditional microscopes (P &lt; 0.05). Preclinical and clinical LED scores did not show a significant difference (P &lt; 0.05). Using a Pearson correlation analysis, a correlation between LED scores and operating time was found. Conclusion: The authors' study’s findings, in conjunction with existing literature, herald a promising shift towards more ergonomic microsurgical practices through the use of robotic assistance and three-dimensional mesoscopic systems. Continued research and innovation in this direction are not only anticipated but necessary for the evolution of microsurgical techniques and the well-being of surgeons in the field.

Irradiation safety, anesthesia time, surgical complications, and patient-reported outcomes in the robotic Mazor X versus fluoroscopy guided minimally invasive transforaminal lumbar interbody fusion surgery: a comparative cohort study.

Robot-assisted (RA) technology is becoming more widely integrated and accepted in spine surgery. The authors sought to evaluate operative and patient-reported outcomes (PROs) in RA versus fluoroscopy-assisted (FA) pedicle screw placement during minimally invasive surgery (MIS) transforaminal lumbar interbody fusion (TLIF). The authors retrospectively studied elective patients who underwent single- or multilevel MIS TLIF for degenerative indication using FA versus RA pedicle screw placement. Patients were selected from September 2021 to May 2023 at a single institution with multiple surgeons whose practice consists of primarily MIS. Outcomes included fluoroscopy dosage per screw, operative time per screw, anesthesia time per screw, estimated blood loss (EBL), screw revision rate, inpatient surgical complications, and minimal clinically important difference (MCID) of Oswestry Disability Index (ODI) and numeric rating scale (NRS) scores at the 6- and 12-month follow-ups. Comparability of groups was analyzed by univariate analysis. Multivariable analysis modeling fluoroscopy time per screw was performed, adjusting for confounders. One hundred eighty-three patients (n = 133 in the FA group vs 50 in the RA group) were included. Patients in the RA cohort were significantly younger than those in the FA group (mean age 63.8 ± 11.9 vs 59.8 ± 11.0 years, p = 0.037). A total of 932 pedicle screws were placed (mean 5.1, range 4-8 per patient). The RA cohort demonstrated significantly lower intraoperative fluoroscopy dosage per screw (4.9 ± 7.6 mGy per screw vs 20.3 ± 14.0 mGy per screw, p < 0.001), significantly longer anesthesia time per screw (49.1 ± 12.6 vs 43.6 ± 9.2, p = 0.009), and similar operative time per screw (33.3 vs 30.7 minutes, p = 0.125). The screw revision rate for symptomatic radiculopathy was zero in both groups. Revision surgery requiring screw removal or reposition was performed in 4 total cases (RA group: 1/50 for infection; FA group: 2/133 for infection, 1/133 for foraminotomy). Both groups demonstrated significant improvement in PROs at 6 and 12 months compared with preoperatively. Moreover, both groups achieved MCID at similar rates. When implementing RA technology, one can expect similar perioperative outcomes as FA techniques in addition to significantly lower radiation exposure. Moreover, there is no statistically significant difference in postoperative PROs between RA and FA. Longer anesthesia times may also be encountered, as in this study, which is likely a result of more complex robotic setup and workflow.

Robotic assistance improves success of cementless component fixation in one total knee arthroplasty system

IntroductionNew cementless implant designs in total knee arthroplasty (TKA) have begun to shift the longstanding practice of cemented fixation. With aseptic loosening a leading cause for revision of cementless implants, initial osteointegration is critical for component survivorship. Robotic-assisted TKA (RA-TKA) has shown promising results in recent literature at improving component accuracy. The current study aims to evaluate if robotic assistance affects the success of cementless fixation in primary TKA. Methods445 cementless primary TKA components from one manufacturer implanted by five surgeons between June 2018 and October 2022 were retrospectively reviewed. Those with minimum one-year follow-up were analyzed. Femoral and tibial components were reviewed separately and grouped based on whether manual or RA-TKA from the same manufacturer was performed. Fisher’s exact test was used to analyze if aseptic loosening rates were different between the two techniques. Results373 (82%) cementless components from a single knee system were included. 146 femoral and 103 tibial components were implanted using RA-TKA, while 63 femoral and 61 tibial components were implanted manually. At a mean follow-up of 18 months (range, 12 to 51 months), successful fixation was achieved in 96.2% of all components. No femoral components from either group were revised due to aseptic loosening. Four manually implanted vs. no robotically assisted tibial components were revised due to aseptic loosening (6.6% vs 0.0%; P = 0.038). ConclusionThe performance of modern cementless femoral components was excellent with or without robotic assistance, however RA-TKA improved the survivorship of the same system’s cementless tibial component.Level of Evidence: Level III.

Autonomy and Use of Flexible Endoscopy Robots in Gastroenterology – A Review

Abstract The integration of mechatronic systems to improve the control and maneuverability of endoscopes in complex gastroenterology procedures is necessary since manual handling often lacks precision. The result are flexible endoscopy robots. Therefore, the question arises as to how far these robots have evolved. This basically depends on their degree of autonomy and the typical interventions they are designed to perform. In this review, we provide an overview of existing flexible endoscopy robotic systems, categorizing their levels of autonomy and intended applications. Of the 28 systems reviewed, most operate at autonomy levels 0 (no autonomy) and level 1 (robotic assistance), with some progress toward level 2 (task automation) in colonoscopy. For more complex procedures, efforts are made to establish master-slave systems. Advances in flexible endoscopic robots are aimed at increasing autonomy, particularly for autonomous navigation within hollow organs. The integration of advanced sensor technologies is critical to achieving precise control and steering in future developments, and thus higher levels of autonomy for these systems.

Design and use of assistant robotic arm in endoscopic transnasal surgery

OBJECTIVE This study aimed to develop a robotic system with autonomous functionality for holding and manipulating the endoscope in endoscopic endonasal surgery. A cost-effective prototype was created with the goal of using robotic assistance and achieving a more efficient learning curve for endoscopic surgery. METHODS A Stewart platform–based endoscope holder and positioner robotic system, developed by the Mechatronics Engineering Department of Kocaeli University, was used in conjunction with 3D Slicer and Plus Toolkit open-source health technology software. Following initial configuration on a 3D skull model, the system was applied to a fresh-frozen cadaver. RESULTS The robotic system and its registration process proved feasible, allowing the robot to autonomously reach predetermined anatomical points during the nasal step of the surgery. Compared with a previous study by the same authors, the surgery duration was successfully shortened. The experiment yielded positive outcomes in both the artificial skull model and the cadaver. The robotic system effectively performed the nasal stage of the surgery. The integration of a robotic surgical arm in the endoscopic endonasal approach holds the potential to enhance the learning curve through shared experiences. The robot offers assistance to surgeons during the learning phase of the endoscopic endonasal approach, aiding in the identification of anatomical landmarks through preoperative planning. CONCLUSIONS This system may present advantages such as improved visualization, enhanced maneuverability, and reduced user fatigue compared with conventional methods. These findings show the potential of robotic assistance in endoscopic surgical techniques.

The safety, accuracy, and feasibility of robotic assistance in neuro-oncological surgery.

Over the past 3 decades, robotic technology has advanced significantly across surgical fields, driven by improvements in versatility, stability, skill, and tactile properties. Neurosurgery has led the way in integrating robotics to improve the accuracy and safety of procedures that require high precision. This study aimed to present one of the largest series in the literature and investigate the feasibility and effectiveness of robotic assistance in neuro-oncological surgery. The authors performed a retrospective review of the medical records of patients who underwent stereotactic biopsy only and biopsy and laser interstitial thermal therapy (LITT) using the Robotic Surgical Assistant (ROSA) system. The ROSA system was used for trajectory planning as well as the precise alignment and insertion of the needle and/or laser catheter. All procedures were performed by a single neurosurgeon. Electronic medical records were reviewed for patient demographics, preoperative clinical deficits, diagnosis, preoperative and postoperative MR images, lesion characteristics including location and volume, postoperative complications, and deficits. A total of 348 patients were identified. The mean age at surgery was 61.4 years, with 171 (49.1%) females. The most common presentations were GBM (40.5%, n = 141), metastasis (16.4%, n = 57), and radiation necrosis (9.8%, n = 34). A total of 189 (54%) patients underwent stereotactic biopsy only, and 159 (46%) underwent biopsy and LITT. The diagnostic yield of the cases was 98.6%, with only 5 cases having inconclusive pathology. Two (0.6%) patients experienced postoperative complications that resolved during follow-up. Ten (2.9%) patients developed asymptomatic subcentimeter hematomas following the procedure that did not require further intervention. No long-term consequences or permanent deficits occurred in any case within a median follow-up duration of 4.4 months (IQR 1-11.2 months). These results indicate that a robot-assisted approach for stereotactic biopsies and LITT is associated with a comparable diagnostic yield and safety profile for frameless and frame-based techniques. Its precision, user-friendly interface, and adaptability contribute to its suitability for such procedures. Future research, especially in long-term results and cost-effectiveness, will be crucial in fully understanding the utility and potential of this technology for broader applications in the field.

Hybrid Brain-Computer Interfacing Paradigm for Assistive Robotics

Hybrid Brain-Computer Interfacing Paradigm for Assistive Robotics

Older People’s Ethical Framing of Autonomy in Relation to Current and Future Consumer Technologies: The Case of Socially Assistive Robots

Older People’s Ethical Framing of Autonomy in Relation to Current and Future Consumer Technologies: The Case of Socially Assistive Robots

Noise reduction in non-invasive brain-computer interfaces for robot control

Abstract To employ a brain-computer interface to control an assistance robot could help disabled people to achieve a minimum degree of autonomy and reduce their dependence on caregivers for simple tasks. In this project, we use a textile, capacitive brain computer interface (BCI) to control an assistant robot. Our BCI is based on Steady State Visual Evoked Potentials (SSVEP), which are a response of the brain to a periodic visual stimulus at a fixed frequency. Since the brain signals are in the microvolt range, the noise reduction of external disturbances has a high relevance. Here, we demonstrate a correlation method for detecting and eliminating phase shifts between different measurements for averaging. This method allows measurements in unshielded environment of two microvolts (uV) with an averaging factor of 100 in our non-invasive BCI set-up.

Speech-Controlled Robot Enabling Cognitive Training and Stimulation in Dementia Prevention for Severely Disabled People

Abstract The current German medical S3-guideline on dementia recommends the use of cognitive training and stimulation, e.g. through shared games, for mild cognitive impairment and mild dementia. The use of cobots, which allow direct human-robot collaboration, enables people with paralyzed upper limbs to actively participate in social activities. This research presents a novel approach through the development of a voice-controlled board game specifically tailored to the inclusion needs of people with severe disabilities. The human voice commands recorded via USB microphones are digitally filtered. Speech recognition of the control commands is performed using a Convolutional Neural Network (CNN) based on the VGG- 16 architecture. The robot´s activity is controlled utilizing the ROS 2 robot operating system. A portable table-based complete system with a 3D-printed Tic-Tac-Toe playing field and robot assistant for severely disabled paralyzed people has been developed. The robot control system employs a pick-and-place mechanism, seamlessly integrating with speech recognition to enhance gameplay interaction. The CNN model achieves an impressive accuracy rate of 97%, ensuring reliable speech recognition performance throughout gameplay. The targeted integration of robotic technologies and artificial intelligence opens new avenues in the prevention of mental illness, care support and inclusion of older and disabled people.

Robot-assisted versus manually guided stereotactic biopsy for intracranial lesions - a systematic review and meta-analysis.

Stereotactic biopsies are essential for obtaining accurate histopathological analysis to guide treatment decisions for deep-seated brain lesions. Manually guided methods, while commonly used and highly precise, can be limited by potential inaccuracies and insufficient tissue sampling. The emerging robot-assisted (RA) techniques offer enhanced precision and could address these limitations for improved accuracy. We aimed to compare safety and effectiveness of RA stereotactic biopsies to traditional techniques. We conducted a systematic search of PubMed, Embase, and Web of Science using terms related to robotic biopsies and intracranial lesions. Eligible studies compared robot-assisted procedures to traditional techniques. Data extracted included diagnostic yield, complication rates, and accuracy. Quality assessment of studies was performed with the ROBINS-I tool, and a random-effects model meta-analysis was performed. Five studies were included, encompassing 913 patients (robot-assisted n = 434, traditional n = 479). Robot-assisted biopsies demonstrated higher diagnostic yield (OR 2.06, 95% CI [1.01,4.21], p = 0.04), and improved trajectory accuracy, with entry point error (EPE) reduced by 0.42mm (95% CI [-0.59,-0.26], p < 0.001) and target point error (TPE) reduced by 0.93mm (95% CI [-1.80,-0.06], p < 0.01). There were no significant differences in the incidence of complications or operative time. RA stereotactic biopsies may improve diagnostic yield and accuracy in targeting intracranial lesions without increasing complication rates or operative time. The limited number of studies and potential biases suggest a need for further high-quality, prospective research to ascertain these findings.

Accuracy of Pedicle Screw Placement Using the ExcelsiusGPS Robotic Navigation Platform: An Analysis of 728 Screws.

Robotic platforms have increased in sophistication for pedicle screw placement. Here, we review our institutional experience using ExcelsiusGPS to assess the accuracy rate of pedicle screw placement throughout the spine and characterize predictors of placement inaccuracy. Retrospective cohort study. Patients from 2017 to 2022 undergoing spinal fusion surgery with ExelsiusGPS-assisted screw implantation at a single tertiary center were retrospectively identified. Patient demographics, preoperative symptoms, and operative details were collected. Postoperative computed tomography was used to classify screw placement accuracy according to the Gertzbein and Robbins scale (GRS). A stepwise multivariable ordered logistic regression analysis determined independent risk factors for clinically inaccurate screws (GRS C/D/E). One hundred and seventeen patients were included. Mean age was 60.6 ± 13.2 years, with 57% men, 72% white, and mean body mass index of 29.9 ± 6.4 kg/m2. Seven hundred and twenty-eight screws were placed, predominantly in the thoracic (29.5%) and lumbar (52.6%) regions. Accuracy classification indicated 670 GRS A, 31 GRS B, 22 GRS C, 4 GRS D, and 1 GRS E screws. The clinically acceptable screw placement rate (GRS A/B) was 96%. Male gender (odds ratio [OR]: 2.12, P = 0.03), revision surgery (OR: 2.43, P = 0.02), and thoracic level screw insertion (OR: 2.33, P = 0.01) were independently associated with inaccurate screw placement and explained 8.7% of the variability seen. Of the 728 screws placed, 3 required revision after postoperative imaging revealed loosening or pedicle breach. ExcelsiusGPS-assisted screw insertion has high placement accuracy and low revision rates. Identification of predictors of inaccuracy illustrates that similar variables, such as placement in the thoracic spine and revision surgery status, apply to both freehand and robotic screw placement. Robotic spine surgery is an accurate, reliable tool that can improve patient outcomes. Factors like male gender, thoracic screw placement, and revision surgery status are associated with lower screw placement accuracy, and these factors should inform surgical decision-making when using robotic assistance.

The role of artificial intelligence in enhancing healthcare for people with disabilities

The integration of artificial intelligence (AI) in healthcare delivery represents a transformative opportunity to enhance the lives of people living with disabilities. AI-driven technologies, such as assistive devices, conversational agents, and rehabilitation tools, can mitigate health disparities, improve diagnostic accuracy, and facilitate effective communication with healthcare providers, fostering more equitable healthcare environments. This commentary explores these applications while addressing the ethical challenges and limitations associated with AI deployment. Specific challenges, such as algorithmic bias, privacy risks with patient data, and the complexity of designing inclusive technologies, are discussed to provide a balanced perspective. For example, biased diagnostic tools may lead to inequitable care, and privacy breaches can compromise sensitive data. Key areas of focus include personalised care through AI-powered systems, the design of inclusive AI technologies incorporating continuous feedback loops and partnerships with advocacy groups, and the development of AI-enabled robotics for physical assistance. This commentary paper emphasises the importance of addressing these limitations alongside advancing ethical AI practices and ensuring continuous user involvement to meet the diverse needs of people living with disabilities, ultimately promoting greater independence and participation in society. Consequently, while AI holds transformative potential in advancing equitable and inclusive healthcare for people with disabilities, addressing ethical challenges, overcoming limitations, and fostering user-centred design are essential to fully realise its benefits and ensure these innovations promote autonomy, accessibility, and well-being.