| 1. |
- Patel, Chirag, et al.
(författare)
-
DBGC : Dimension-Based Generic Convolution Block for Object Recognition
- 2022
-
Ingår i: Sensors. - : MDPI. - 1424-8220. ; 22:5
-
Tidskriftsartikel (refereegranskat)abstract
- The object recognition concept is being widely used a result of increasing CCTV surveillance and the need for automatic object or activity detection from images or video. Increases in the use of various sensor networks have also raised the need of lightweight process frameworks. Much research has been carried out in this area, but the research scope is colossal as it deals with open-ended problems such as being able to achieve high accuracy in little time using lightweight process frameworks. Convolution Neural Networks and their variants are widely used in various computer vision activities, but most of the architectures of CNN are application-specific. There is always a need for generic architectures with better performance. This paper introduces the Dimension-Based Generic Convolution Block (DBGC), which can be used with any CNN to make the architecture generic and provide a dimension-wise selection of various height, width, and depth kernels. This single unit which uses the separable convolution concept provides multiple combinations using various dimension-based kernels. This single unit can be used for height-based, width-based, or depth-based dimensions; the same unit can even be used for height and width, width and depth, and depth and height dimensions. It can also be used for combinations involving all three dimensions of height, width, and depth. The main novelty of DBGC lies in the dimension selector block included in the proposed architecture. Proposed unoptimized kernel dimensions reduce FLOPs by around one third and also reduce the accuracy by around one half; semi-optimized kernel dimensions yield almost the same or higher accuracy with half the FLOPs of the original architecture, while optimized kernel dimensions provide 5 to 6% higher accuracy with around a 10 M reduction in FLOPs.
|
|
| 2. |
- Poveda, Alaitz, et al.
(författare)
-
Exposome-wide ranking of modifiable risk factors for cardiometabolic disease traits
- 2022
-
Ingår i: Scientific Reports. - : Nature Publishing Group. - 2045-2322. ; 12:1
-
Tidskriftsartikel (refereegranskat)abstract
- The present study assessed the temporal associations of ~ 300 lifestyle exposures with nine cardiometabolic traits to identify exposures/exposure groups that might inform lifestyle interventions for the reduction of cardiometabolic disease risk. The analyses were undertaken in a longitudinal sample comprising > 31,000 adults living in northern Sweden. Linear mixed models were used to assess the average associations of lifestyle exposures and linear regression models were used to test associations with 10-year change in the cardiometabolic traits. ‘Physical activity’ and ‘General Health’ were the exposure categories containing the highest number of ‘tentative signals’ in analyses assessing the average association of lifestyle variables, while ‘Tobacco use’ was the top category for the 10-year change association analyses. Eleven modifiable variables showed a consistent average association among the majority of cardiometabolic traits. These variables belonged to the domains: (i) Smoking, (ii) Beverage (filtered coffee), (iii) physical activity, (iv) alcohol intake, and (v) specific variables related to Nordic lifestyle (hunting/fishing during leisure time and boiled coffee consumption). We used an agnostic, data-driven approach to assess a wide range of established and novel risk factors for cardiometabolic disease. Our findings highlight key variables, along with their respective effect estimates, that might be prioritised for subsequent prediction models and lifestyle interventions.
|
|
