Automated detection of anterior cruciate ligament tears using a deep convolutional neural network.

BACKGROUND: The development of computer-assisted technologies to diagnose anterior cruciate ligament (ACL) injury by analyzing knee magnetic resonance images (MRI) would be beneficial, and convolutional neural network (CNN)-based deep learning approaches may offer a solution. This study aimed to evaluate the accuracy of a CNN system in diagnosing ACL ruptures by a single slice from a knee MRI and to compare the results with that of experienced human readers. METHODS: One hundred sagittal MR images from patients with and without ACL injuries, confirmed by arthroscopy, were cropped and used for the CNN training. The final decision by the CNN for intact or torn ACL was based on the probability of ACL tear on a single MRI slice. Twelve board-certified physicians reviewed the same images used by CNN. RESULTS: The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of the CNN classification was 91.0%, 86.0%, 88.5%, 87.0%, and 91.0%, respectively. The overall values of the physicians' readings were similar, but the specificity was lower than the CNN classification for some of the physicians, thus resulting in lower accuracy for the human readers. CONCLUSIONS: The trained CNN automatically detected the ACL tears with acceptable accuracy comparable to that of human readers.

Therapeutic effects of sleeve gastrectomy for non-alcoholic steatohepatitis estimated by paired liver biopsy in morbidly obese Japanese patients.

ABSTRACT: Bariatric surgery has been reported to improve non-alcoholic steatohepatitis (NASH), which is a frequent comorbidity in morbidly obese patients. We performed a retrospective cohort study to estimate the therapeutic effect of sleeve gastrectomy (SG), the most common bariatric surgery in Japan, on obese patients with NASH by comparing the findings of paired liver biopsies.Eleven patients who underwent laparoscopic SG for the treatment of morbid obesity, defined as body mass index (BMI) > 35 kg/m2, from March 2015 to June 2019 at Hiroshima University Hospital, Japan, were enrolled. All patients were diagnosed with NASH by liver biopsy before or during SG and were re-examined with a second liver biopsy 1 year after SG. The clinical and histological characteristics were retrospectively analyzed.One year after SG, body weight and BMI were significantly reduced, with median reductions in body weight and BMI of-22 kg and -7.9 kg/m2, respectively. Body fat was also significantly reduced at a median of 13.7%. Liver-related enzymes were also significantly improved. On re-examination by paired liver biopsy, liver steatosis improved in 9 of the 11 patients (81.8%), ruling out of the pathological diagnosis of NASH. However, fibrosis stage did not significantly improve 1 year after SG. The non-alcoholic fatty liver disease activity score was significantly reduced in 10 of 11 patients (90.9%).Pathological improvement or remission of NASH could be achieved in most morbidly obese Japanese patients 1 year after SG.

Possible role of MRI-detected osteophytes as a predictive biomarker for development of osteoarthritis of the knee: A study using data from the Osteoarthritis Initiative.

Objective: To elucidate the possible role of MRI-detected osteophytes as a predictive imaging biomarker for knee osteoarthritis (KOA). Design: Subjects (n â€‹= â€‹303) were selected according to the following inclusion criteria from the Osteoarthritis Initiative (OAI) data set: (1) â€‹< â€‹55 years old; (2) Western Ontario and McMaster Universities Arthritis Index pain score of 0; (3) Kellgren-Lawrence (KL) system grade 0 or 1; and (4) Complete MRI data set of the right knee. A pre-OA group (POA) consisted of subjects who developed KL grade 2 or more within 96 months, and a non-OA group (NOA) that remained KL 0 or 1 during that period. Baseline MRIs were assessed for osteophyte formation. Twenty-five locations were examined according to the MOAKS osteophyte score. Osteophytes at each location were assessed in terms of their predictive value for OA development. Results: Thirty-two subjects were POA and 271 were NOA. Age, BMI, and sex did not differ between the two groups. In the POA group, the number of subjects with osteophytes tended to be higher at all 25 sites. Forward stepwise regression analysis revealed five locations - medial patella, lateral intra-condylar notch of the femur, lateral femoral condyle, tibial spine, and lateral posterior condyle - were important for the prediction of KOA development. Having more than two osteophytes at these five locations predicted KOA development with a sensitivity of 0.75 and specificity of 0.79. Conclusions: MRI-detected osteophytes could serve as a predictive biomarker of KOA development within 96 months after detection.