Background: Artificial intelligence (AI) has gained tremendous popularity recently, especially the use of natural language processing (NLP). ChatGPT is a state-of-the-art chatbot capable of creating natural conversations using NLP. The use of AI in medicine can have a tremendous impact on health care delivery. Although some studies have evaluated ChatGPT’s accuracy in self-diagnosis, there is no research regarding its precision and the degree to which it recommends medical consultations. Objective: The aim of this study was to evaluate ChatGPT’s ability to accurately and precisely self-diagnose common orthopedic diseases, as well as the degree of recommendation it provides for medical consultations. Methods: Over a 5-day course, each of the study authors submitted the same questions to ChatGPT. The conditions evaluated were carpal tunnel syndrome (CTS), cervical myelopathy (CM), lumbar spinal stenosis (LSS), knee osteoarthritis (KOA), and hip osteoarthritis (HOA). Answers were categorized as either correct, partially correct, incorrect, or a differential diagnosis. The percentage of correct answers and reproducibility were calculated. The reproducibility between days and raters were calculated using the Fleiss κ coefficient. Answers that recommended that the patient seek medical attention were recategorized according to the strength of the recommendation as defined by the study. Results: The ratios of correct answers were 25/25, 1/25, 24/25, 16/25, and 17/25 for CTS, CM, LSS, KOA, and HOA, respectively. The ratios of incorrect answers were 23/25 for CM and 0/25 for all other conditions. The reproducibility between days was 1.0, 0.15, 0.7, 0.6, and 0.6 for CTS, CM, LSS, KOA, and HOA, respectively. The reproducibility between raters was 1.0, 0.1, 0.64, –0.12, and 0.04 for CTS, CM, LSS, KOA, and HOA, respectively. Among the answers recommending medical attention, the phrases “essential,” “recommended,” “best,” and “important” were used. Specifically, “essential” occurred in 4 out of 125, “recommended” in 12 out of 125, “best” in 6 out of 125, and “important” in 94 out of 125 answers. Additionally, 7 out of the 125 answers did not include a recommendation to seek medical attention. Conclusions: The accuracy and reproducibility of ChatGPT to self-diagnose five common orthopedic conditions were inconsistent. The accuracy could potentially be improved by adding symptoms that could easily identify a specific location. Only a few answers were accompanied by a strong recommendation to seek medical attention according to our study standards. Although ChatGPT could serve as a potential first step in accessing care, we found variability in accurate self-diagnosis. Given the risk of harm with self-diagnosis without medical follow-up, it would be prudent for an NLP to include clear language alerting patients to seek expert medical opinions. We hope to shed further light on the use of AI in a future clinical study.

Thumb opposition is vital for hand function and involves pronation and palmar abduction. The improvement of pronation is often used as one of the evaluation items of the opponensplasty method for severe carpal tunnel syndrome. However, most of the studies used substitution evaluation methods for measurement of the pronation angle. Thus, there is still no appropriate method for measuring thumb pronation angle accurately in carpal tunnel syndrome patients. In recent reports, a wearable gyroscope was used to evaluate upper extremity motions and it can be possibly used for accurate measurement of the thumb pronation angle along the three-dimensionally moving bone axis. Thus, we investigated the reliability of measuring thumb pronation using a gyroscope and evaluated whether this method can be used to detect opposition impairment. The participants were volunteers with unaffected upper limbs (32 hands) and patients with carpal tunnel syndrome (27 hands). The pronation and palmar abduction angles during opposition movements were measured using a three-axis gyroscope that included a three-axis accelerometer. The gyroscope was fixed onto the first metacarpal bone and the thumb phalanx. The pronation and palmar abduction angles of the metacarpal bone and the palmar abduction angles of the phalanx significantly decreased in the carpal tunnel syndrome group. The pronation angle of the metacarpal bone during opposition movement peaked later than the palmar abduction angle in all hands. We were able to measure the thumb pronation and palmar abduction angles using the three-axis gyroscope, and this tool was able to detect impairments of thumb opposition due to carpal tunnel syndrome. This could be a tool for measuring thumb and finger angles and for detecting impairments caused by various diseases.

Background: Carpal tunnel syndrome (CTS), the most common neuropathy, is caused by a compression of the median nerve in the carpal tunnel and is related to aging. The initial symptom is numbness and pain of the median nerve distributed in the hand area, while thenar muscle atrophy occurs in advanced stages. This atrophy causes failure of thumb motion and results in clumsiness; even after surgery, thenar atrophy does not recover for an extended period. Medical examination and electrophysiological testing are useful to diagnose CTS; however, visits to the doctor tend to be delayed because patients neglect the symptom of numbness in the hand. To avoid thenar atrophy-related clumsiness, early detection of CTS is important. Objective: To establish a CTS screening system without medical examination, we have developed a tablet-based CTS detection system, focusing on movement of the thumb in CTS patients; we examined the accuracy of this screening system. Methods: A total of 22 female CTS patients, involving 29 hands, and 11 female non-CTS participants were recruited. The diagnosis of CTS was made by hand surgeons based on electrophysiological testing. We developed an iPad-based app that recorded the speed and timing of thumb movements while playing a short game. A support vector machine (SVM) learning algorithm was then used by comparing the thumb movements in each direction among CTS and non-CTS groups with leave-one-out cross-validation; with this, we conducted screening for CTS in real time. Results: The maximum speed of thumb movements between CTS and non-CTS groups in each direction did not show any statistically significant difference. The CTS group showed significantly slower average thumb movement speed in the 3 and 6 o’clock directions (P=.03 and P=.005, respectively). The CTS group also took a significantly longer time to reach the points in the 2, 3, 4, 5, 6, 8, 9, and 11 o’clock directions (P<.05). Cross-validation revealed that 27 of 29 CTS hands (93%) were classified as having CTS, while 2 of 29 CTS hands (7%) did not have CTS. CTS and non-CTS were classified with 93% sensitivity and 73% specificity. Conclusions: Our newly developed app could classify disturbance of thumb opposition movement and could be useful as a screening test for CTS patients. Outside of the clinic, this app might be able to detect middle-to-severe-stage CTS and prompt these patients to visit a hand surgery specialist; this may also lead to medical cost-savings.

The Kapandji score is used widely to record thumb opposition and is the only alternative to angular measurements. We aimed to verify the extent to which the Kapandji score reflects the angle of pronation. We recruited volunteers with unaffected upper limbs (33 hands) and patients with carpal tunnel syndrome (20 hands). We attached a small gyroscope to the thumb metacarpal and measured the pronation and palmar abduction angles at each position of the Kapandji score in both groups. The Kapandji score showed a strong ceiling effect, and the increase in angle size at each Kapandji score levelled off around a score of 6 in both groups. A Kapandji score greater than 6 does not necessarily reflect thumb opposition. Level of Evidence: III