ACCURACY OF ARTIFICIAL INTELLIGENCE AS CLINICAL DECISION SUPPORT SYSTEM IN DIAGNOSING CERVICAL RADICULOPATHY DUE TO DISC HERNIATION AND SPONDYLOSIS

M.F. Chevidikunnan1, A.K. Alzahrani1, U.M. Alabasi1, F.R. Khan1, F.H. Badahman1, S. Thomas2
1King Abdulaziz University, Physical Therapy, Faculty of Medical Rehabilitation Sciences, Jeddah, Saudi Arabia, 2THERAPHA Inc, New York, United States

Background: Neck pain is one of the most common musculoskeletal disorders in adults worldwide, with prevalence ranging from 17% up to 75%. Even in Saudi Arabia the prevalence ranging from 33% up to 85%. In specific, cervical radiculopathy is a neurological condition caused by malfunctioning of nerve roots in the cervical spine region. CT or MRI are the widespread way to diagnose these problems radiologically, which are found to be expensive for many patients. Meanwhile, others may suffer from other health diseases that prevent them from undergoing these radiations, which may affect the appropriate diagnosis of the condition. On the other hand, Artificial Intelligence (AI) is the problem- solving process that depends on the progression and modulation of computer systems combined with human intelligence. AI can improve healthcare by advancing the decision- making process, care delivery, and patient participation, thereby the artificial intelligence could be appropriate, accurate and suitable models to diagnose cervical radiculopathy.

Purpose: The study aim to compare the accuracy of an AI-enabled platform and an Algorithm as a Clinical Decision Support System (CDSS) versus MRI and CT scan in diagnosing patients affected with cervical disc herniation and spondylosis.

Methods:
It was a Cross-sectional study, where fifty-five male and female patients above 18 year of age, who suffer from neck pain were included in the study. the patients were excluded if they had a previous history of cervical disc herniation, cancer, patients who prohibited from radio-diagnostic procedures, and uncooperative patients. The personal and clinical history was taken by using a newly developed and patented software Therapha™on the same day or 2 to 3 days before or after the patient undergoes a MRI or CT scan. The diagnostic accuracy of AI was determined in terms of sensitivity and specificity compared with MRI or CT Scan.

Results:
The data were analyzed by using the SPSS software version 23.0. The results showed that, AI based Therapha™ software has a high level of sensitivity (98%), and specificity (63%). It also showed, a positive predictive value of 94%, and a negative predictive value of 83.3%. The area under the receiving operating curve (ROC) was 0.80.

Conclusions:
The study results conclude that, the Therapha™ Software showed a high level of sensitivity and specificity for the diagnosis of cervical radiculopathy, so thereby it could be of a useful tool for the Rehabilitation Practitioner, which provides the diagnosis according to the physical examination of the patients.

Implications: The AI based Therapha™ software could be used as a tool to diagnose cervical radiculopathy, which could be highly recommended in rehabilitation centers where the highly sophisticated radio- diagnostic facilities are unavailable.

Funding acknowledgements: Nil

Keywords:
Artificial Intelligence
Cervical Radiculopathy
Diagnostic Accuracy

Topics:
Innovative technology: information management, big data and artificial intelligence
Musculoskeletal: spine
Musculoskeletal

Did this work require ethics approval? Yes
Institution: King Abdulaziz University Hospital (KAUH), Jeddah, Saudi Arabia
Committee: Institutional Review Board, King Abdulaziz University Hospital, Jeddah, Saudi Arabia
Ethics number: HA-02-J-008.

All authors, affiliations and abstracts have been published as submitted.

Back to the listing