Eight pruning deep learning models for low storage and high-speed COVID-19 computed tomography lung segmentation and heatmap-based lesion localization: A multicenter study using COVLIAS 2.0.
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Agarwal Mohit, Agarwal Sushant, Saba Luca, Chabert Gian Luca, Gupta Suneet, Carriero Alessandro, Pasche Alessio, Danna Pietro, Mehmedovic Armin, Faa Gavino, Shrivastava Saurabh, Jain Kanishka, Jain Harsh, Jujaray Tanay, Singh Inder M, Turk Monika, Chadha Paramjit S, Johri Amer M, Khanna Narendra N, Mavrogeni Sophie, Laird John R, Sobel David W, Miner Martin, Balestrieri Antonella, Sfikakis Petros P, Tsoulfas George, Misra Durga Prasanna, Agarwal Vikas, Kitas George D, Teji Jagjit S, Al-Maini Mustafa, Dhanjil Surinder K, Nicolaides Andrew, Sharma Aditya, Rathore Vijay, Fatemi Mostafa, Alizad Azra, Krishnan Pudukode R, Yadav Rajanikant R, Nagy Frence, Kincses Zsigmond Tamás, Ruzsa Zoltan, Naidu Subbaram, Viskovic Klaudija, Kalra Manudeep K, Suri Jasjit S
Аннотация
METHOD: ology: The proposed study uses multicenter ∼9,000 CT slices from two different nations, namely, CroMed from Croatia (80 patients, experimental data), and NovMed from Italy (72 patients, validation data). We hypothesize that by using pruning and evolutionary optimization algorithms, the size of the AI models can be reduced significantly, ensuring optimal performance. Eight different pruning techniques (i) differential evolution (DE), (ii) genetic algorithm (GA), (iii) particle swarm optimization algorithm (PSO), and (iv) whale optimization algorithm (WO) in two deep learning frameworks (i) Fully connected network (FCN) and (ii) SegNet were designed. COVLIAS 2.0 was validated using "Unseen NovMed" and benchmarked against MedSeg. Statistical tests for stability and reliability were also conducted.
RESULTS: Pruning algorithms (i) FCN-DE, (ii) FCN-GA, (iii) FCN-PSO, and (iv) FCN-WO showed improvement in storage by 92.4%, 95.3%, 98.7%, and 99.8% respectively when compared against solo FCN, and (v) SegNet-DE, (vi) SegNet-GA, (vii) SegNet-PSO, and (viii) SegNet-WO showed improvement by 97.1%, 97.9%, 98.8%, and 99.2% respectively when compared against solo SegNet. AUC > 0.94 (p < 0.0001) on CroMed and > 0.86 (p < 0.0001) on NovMed data set for all eight EA model. PAI <0.25 s per image. DenseNet-121-based Grad-CAM heatmaps showed validation on glass ground opacity lesions.
CONCLUSIONS: Eight PAI networks that were successfully validated are five times faster, storage efficient, and could be used in clinical settings.