SleepNet: automated sleep analysis via dense convolutional neural network using physiological time series.

OBJECTIVE: In this work, a dense recurrent convolutional neural network (DRCNN) was constructed to detect sleep disorders including arousal, apnea and hypopnea using polysomnography (PSG) measurement channels provided in the 2018 PhysioNet Challenge database. APPROACH: Our model structure is composed of multiple dense convolutional units (DCU) followed by a bidirectional long-short term memory (LSTM) layer followed by a softmax output layer. The sleep events, including sleep stages, arousal regions and multiple types of apnea and hypopnea, are manually annotated by experts, which enables us to train our proposed network using a multi-task learning mechanism. Three binary cross-entropy loss functions, corresponding to sleep/wake, target arousal and apnea-hypopnea/normal detection tasks, are summed up to generate our overall network loss function that is optimized using the Adam method. Our model performance was evaluated using two metrics: the area under the precision-recall curve (AUPRC) and the area under the receiver operating characteristic curve (AUROC). To measure our model generalization, 4-fold cross-validation was also performed. For training, our model was applied to full night recording data. MAIN RESULTS: Finally, the average AUPRC and AUROC values associated with the arousal detection task were 0.505 and 0.922, respectively, on our testing dataset. An ensemble of four models trained on different data folds improved the AUPRC and AUROC to 0.543 and 0.931, respectively. SIGNIFICANCE: Our proposed algorithm achieved the first place in the official stage of the 2018 PhysioNet Challenge for detecting sleep arousals with an AUPRC of 0.54 on the blind testing dataset.

Arthroscopic washout of the knee--a 5-year survival analysis.

Arthroscopy and washout of the knee is commonly performed for early osteoarthritis. Very little information exists regarding long-term prognosis, especially in terms of avoidance of further surgery. Using a prospectively gathered database, 100 consecutive patients having knee arthroscopy with a finding of OA between 1991 and 1993 were identified and their outcome at 5 years ascertained. Ninety-nine patients with 100 arthroscoped knees were identified. Fifty-eight had isolated medial compartment disease and six had isolated lateral compartment disease. In 36, both compartments were affected. Eighteen knees had further major surgery during follow up; 11 had total knee replacement, four had high tibial osteotomy and three had unicondylar knee arthroplasty. Those requiring surgery were significantly older (62 cf. 53 years, P=0.008). Meniscectomy was not an important risk factor (chi2, P=0.67). The rate of knee survival without operation at 5 years was much lower in those aged over 60 years than in those younger (68% cf. 89%). (chi2, P=0.02). Only 18% of patients progress to major knee surgery within 5 years of arthroscopic washout for osteoarthritis. Age greater than 60 years worsens the prognosis considerably.

Tendo Achillis rupture; surgical repair is a safe option.

We reviewed the results of 67 tendo Achillis repairs performed at one District General Hospital over a 5-year period, using details from a comprehensive database. At a mean follow-up of 2.5 years, the re-rupture rate was 3% (2 out of 67). The infection rate was 1.5% (1 out of 67). A review of the relevant literature has revealed that quoted infection rates following tendo Achillis repair vary considerably, and are as high as 21%. Several authors have advised against repair based on those data. In the light of our results, we can recommend surgical repair of tendo Achillis rupture as a safe option.

Pain relief following tennis elbow release.

We studied the outcome of tennis elbow release in 27 patients at an average of 29.6 months after surgery. We found that 44% of patients had obtained complete pain relief, 37% of patients experienced occasional pain and 19% of patients still experienced moderate pain. Pain relief was significantly better in those patients with the shorter duration of preoperative symptoms. We therefore conclude that surgery for tennis elbow should be employed at an earlier stage than is currently practised.

Free vascularised epiphyseal transplantation in upper extremity reconstruction.

We describe preliminary observations on the use of free vascularised transfer of proximal fibular epiphysis in replacement of proximal humeral epiphysis or distal radial epiphysis in three patients. An analysis of the early results shows evidence of longitudinal growth, graft hypertrophy and epiphyseal adaptation in transplanted epiphysis. A review of the problems in this type of surgery is included.

A gastrocnemius-pedicled femoral bone graft in resection arthrodesis at the knee.

We describe a new method of reconstruction after resection of tumours of the proximal tibia by grafting and arthrodesis of the knee. Two separate vascularised bone grafts from the ipsilateral limb were used, one a gastrocnemius-pedicled femoral graft and the other a pedicled fibular graft. An anatomical study of the gastrocnemius-pedicled femoral graft was made. The method was shown to be practical and reproducible. One patient with osteosarcoma has a successful result with no recurrence at two-and-a-half years.