Perioperative management of diabetes in elective patients: a region-wide audit.

BACKGROUND: Ten percent of elective surgical patients have diabetes. These patients demonstrate excess perioperative morbidity and mortality. National guidance on the management of adults with diabetes undergoing surgery was published in 2011. We present a region-wide audit of adherence to this guidance across the North Western Deanery. METHODS: Local teams prospectively collected data according to a locally approved protocol. Pregnant, paediatric and non-elective patients were excluded from this audit. Patient characteristics, type of surgery and aspects of perioperative management were collated and centrally analysed against audit criteria based upon national recommendations. RESULTS: 247 patients with diabetes were identified. HbA1c was recorded in 71% of patients preoperatively; 9% of patients with an abnormal HbA1c were not known by, or referred to, the diabetes team. 17% of patients were admitted the evening preceding surgery. The mean fasting time was 12:20(4) h. Variable rate i.v. insulin infusions (VRIII) were not used when indicated in 11%. Only 8% of patients received the recommended substrate fluid, along with the VRIII (5% glucose in 0.45% saline). Intra-operative capillary blood glucose (CBG) was measured hourly in 56% of patients. Intra-operative CBG was within the acceptable range (4-12 mmol.L(-1)) in 85% of patients. 73% of patients had a CBG measurement performed in recovery. The WHO checklist was used in 95% of patients. CONCLUSIONS: National perioperative guidelines were not adhered to in a substantial proportion of patients with diabetes undergoing elective surgery. This study represents a template for future trainee networks.

Content-based image retrieval of Indian traditional textile motifs using deep feature fusion.

In the fast-paced fashion world, unique designs are like early birds, grabbing attention as online shopping surges. Fabric texture plays an immense role in selecting the perfect design. Indian Traditional textile motifs are pivotal, showing rich cultural origins and attracting worldwide art fanatics. Yet, technology-driven abstract forms are posing a challenge for them. The decline of handmade artistic ability due to computerization is concerning. Crafting new designs associated with the latest trends is time- consuming and requires diligence. In this work an interactive CBIR (content-based image retrieval) system is presented. It utilizes deep features from InceptionV3 and InceptionResNetV2 models to match query designs with a database of traditional Indian textiles. Its performance is tested with Caltech-101, Corel-1K state-of-the-art datasets, and Indian Textiles datasets and the results are shown to be finer than the existing approaches. The similarity-based fine-grained saliency maps (SBFGSM) approach is employed to visualize the importance of features. Our approach combines deep feature fusion with PCA dimensionality reduction and speeds up search using a clustering approach. Relevance feedback is employed to refine the retrievals. This tool is expected to benefit designers by accelerating the design cycles by bridging the gap between human creativity and A.I. assistance.