A Food Recognition System for Diabetic Patients based on an Optimized Bag of Features Model

Computer vision-based food recognition could be used to estimate a meal's carbohydrate content for diabetic patients. This study proposes a methodology for automatic food recognition, based on the Bag of Features (BoF) model. An extensive technical investigation was conducted for the identification and optimization of the best performing components involved in the BoF architecture, as well as the estimation of the corresponding parameters. For the design and evaluation of the prototype system, a visual dataset with nearly 5,000 food images was created and organized into 11 classes. The optimized system computes dense local features, using the scale-invariant feature transform on the HSV color space, builds a visual dictionary of 10,000 visual words by using the hierarchical k-means clustering and finally classifies the food images with a linear support vector machine classifier. The system achieved classification accuracy of the order of 78%, thus proving the feasibility of the proposed approach in a very challenging image dataset.

Intra-abdominal pressure development after different temporary abdominal closure techniques in a porcine model

BACKGROUND: Decompressive laparotomy followed by temporary abdominal closure (TAC) is an established prophylaxis and treatment for abdominal compartment syndrome. The herein presented study aimed at the comparison of volume reserve capacity and development of intra-abdominal hypertension after forced primary abdominal closure and different TAC techniques in a porcine model. METHODS: Eight anesthesized and mechanically ventilated domestic pigs underwent a standardized midline laparotomy. A bag was placed into the abdominal cavity. Before abdominal closure, the bag was prefilled with 3,000 mL water to simulate increased intra-abdominal volume. The intra-abdominal pressure (IAP) was then increased in 2 mm Hg steps up to 30 mm Hg by adding volume (volume reserve capacity) to the intra-abdominal bag. Volume reserve capacity with the corresponding IAP were analyzed and compared for primary abdominal closure, bag silo closure, a zipper system, and vacuum-assisted closure (VAC) with different negative pressures (-50, -100, and -150 mm Hg). Hemodynamic and pulmonary parameters were monitored throughout the experiment. RESULTS: Volume reserve capacity was the highest for bag silo closure followed by the zipper system and VAC with primary abdominal closure providing the least volume reserve capacity in the whole IAP range. Of interest, VAC -50 mm Hg resulted in a lower volume reserve capacity when compared with VAC -100 and -150 mm Hg. Pulmonary and hemodynamic parameters demonstrated no significant differences between primary abdominal closure and the evaluated TAC techniques at all IAP levels. CONCLUSIONS: The present experimental in vivo study indicates that bag silo closure and zipper systems may be favorable TAC techniques after decompressive laparotomy. In contrast, the VAC techniques resulted in lower volume reserve capacity and therefore may bear an increased risk for recurrent intra-abdominal hypertension in the initial phase after decompressive laparotomy.