Artificial intelligence and skin melanoma.

Melanoma is the deadliest skin cancer, presenting typically with changing pigmented areas and usually treated with surgical removal. As benign cutaneous pigmented lesions are very common in all populations, it can be challenging to identify which areas should be cut out or left untreated. Delayed treatment in melanoma increases the risk of death, but it is not possible to remove all lesions. Dermatoscopy uses polarized light and can be used to help distinguish melanomas from benign lesions. Dermatoscopy images with a confirmed diagnosis can be used to develop artificial intelligence (AI) as a medical device (AIaMD) tool. This contribution discusses the utilization of AI in melanoma management and describes an AIaMD tool used in current UK clinical practice on more than 80,000 patients. This is a springboard for discussing the scope, risks, and mitigations for future AI use by all clinicians involved in managing people with melanoma.

Validation of a modified-composite autonomic symptom score (COMPASS-31) as an outcome measure for persistent autonomic symptoms post-concussion: an observational pilot study.

OBJECTIVE: Persistent symptoms post-mild traumatic brain injury (mTBI) includes autonomic dysregulation (AD). The composite autonomic symptoms score, (COMPASS-31), was developed to quantify AD symptom severity in the last year, which limits clinical utility. The primary aim was to determine validity of a modified-COMPASS-31 measuring symptoms in the last month compared to the original, secondarily to compare both original and modified versions to the Neurobehavioral Symptom Inventory (NSI), and tertiarily to detect change post-treatment of the modified-COMPASS-31 compared to NSI and headache intensity (HI). PARTICIPANTS: Thirty-three military personnel with persistent headache post-mTBI. MAIN OUTCOME MEASURES: Total and domain scores for COMPASS-31 (original vs. modified) NSI and HI at baseline. Change in modified-COMPASS-31. NSI, and HI. RESULTS: Baseline COMPASS-31 versions were comparable and highly correlated (r = 0.72, p < 0.001), they were moderately correlated at best to the NSI (r < 0.6), which may suggest differences in measurement metrics. The mean change in modified-COMPASS-31 scores (15.4/100, effect size 0.8) was mild to moderately correlated to the change in HI (r = 0.39) score, but not to NSI (r = 0.28). CONCLUSION: The modified-COMPASS-31 appears to be valid, can measure change of AD symptom severity, and is recommended as an outcome measure.

Accuracy of an artificial intelligence as a medical device as part of a UK-based skin cancer teledermatology service.

Introduction: An artificial intelligence as a medical device (AIaMD), built on convolutional neural networks, has demonstrated high sensitivity for melanoma. To be of clinical value, it needs to safely reduce referral rates. The primary objective of this study was to demonstrate that the AIaMD had a higher rate of correctly classifying lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatology standard of care (SoC), while achieving the same sensitivity to detect malignancy. Secondary endpoints included the sensitivity, specificity, positive and negative predictive values, and number needed to biopsy to identify one case of melanoma or squamous cell carcinoma (SCC) by both the AIaMD and SoC. Methods: This prospective, single-centre, single-arm, masked, non-inferiority, adaptive, group sequential design trial recruited patients referred to a teledermatology cancer pathway (clinicaltrials.gov NCT04123678). Additional dermoscopic images of each suspicious lesion were taken using a smartphone with a dermoscopic lens attachment. The images were assessed independently by a consultant dermatologist and the AIaMD. The outputs were compared with the final histological or clinical diagnosis. Results: A total of 700 patients with 867 lesions were recruited, of which 622 participants with 789 lesions were included in the per-protocol (PP) population. In total, 63.3% of PP participants were female; 89.0% identified as white, and the median age was 51 (range 18-95); and all Fitzpatrick skin types were represented including 25/622 (4.0%) type IV-VI skin. A total of 67 malignant lesions were identified, including 8 diagnosed as melanoma. The AIaMD sensitivity was set at 91 and 92.5%, to match the literature-defined clinician sensitivity (91.46%) as closely as possible. In both settings, the AIaMD identified had a significantly higher rate of identifying lesions that did not need a biopsy or urgent referral compared to SoC (p-value = 0.001) with comparable sensitivity for skin cancer. Discussion: The AIaMD identified significantly more lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatologists. This has the potential to reduce the burden of unnecessary referrals when used as part of a teledermatology service.

Is proprioception affected in those with persistent intra-articular temporomandibular disorders? A cross-sectional study exploring joint position sense and force sense of the jaw.

BACKGROUND: Proprioception is an important component of normal jaw function, although to date it is unknown if and how proprioception is affected in those with intra-articular temporomandibular disorders (IA-TMDs). This challenges effective management of this subgroup. OBJECTIVES: Determine whether differences in local joint position sense and force sense exist between individuals with IA-TMD and those without a jaw problem. DESIGN: Cross-sectional study. METHOD: Sixty age and sex matched participants were recruited (n = 30 IA-TMD, n = 30 healthy controls). Temporomandibular joint position sense was tested to one target (50% of maximal mouth opening) using a ruler. Temporomandibular force sense was tested to two targets (50% and 70% of maximal molar pain-free bite force) using a bite sensor. Constant, absolute and variable errors were calculated for each outcome and compared between groups using p-values and 95% confidence intervals for effect size (d). RESULTS: Significantly greater degrees of constant, absolute and variable error were seen for force sense testing at the 50% target (p < 0.05). No statistically significant difference was observed between groups for joint position sense, or for constant or variable force sense error at the 70% target (p > 0.05). Despite this, all joint position sense and force sense measures demonstrated possible clinical significance (upper limits 95% CI d ≥ 0.5). CONCLUSIONS: Force sense, particularly at lower levels, appears impaired in those with IA-TMD. Joint position sense and force sense at higher levels appear less affected, although may still be important to consider in the management of IA-TMD in some patients.