Smartphone-based machine learning algorithm for cervical myelopathy screening with the 10-s grip-and-release test: a pilot study (Preprint)

BACKGROUND As cervical myelopathy (CM) is a progressive disease, early detection and intervention are essential for its mitigation. While several screening methods exist, they are difficult to understand for community-dwelling people, and the equipment required to set up the test environments is expensive. Thus, a simple screening system is necessary to ensure early consultation by a physician. OBJECTIVE This study investigated the viability of a CM screening method based on the 10-s grip-and-release test using a machine learning algorithm and a smartphone equipped with a camera. METHODS The group of CM patients and the control group consisted of 22 and 17 participants, respectively. A spine surgeon diagnosed the presence of CM. Patients performing the 10-s grip-and-release test were filmed, and the videos were analyzed. The probability of the presence of CM was estimated using a support vector machine algorithm, and the sensitivity, specificity, and area under the curve (AUC) were calculated. Two assessments of the correlation between estimated scores were conducted. The first used a random forest regression model and the Japanese Orthopaedic Association Cervical Myelopathy Evaluation Questionnaire (C-JOA). The second assessment used a different model, random forest regression, and the Disabilities of the Arm, Shoulder, and Hand (DASH). RESULTS The final classification model had a sensitivity of 90.9%, specificity of 88.2%, and AUC of 0.93. The correlations between each estimated score and the C-JOA and DASH scores were 0.79 and 0.67, respectively. CONCLUSIONS The proposed model could be a helpful screening tool for CM as it showed excellent performance and high usability for community-dwelling people and non-spine surgeons.