Ying-Hwey Nai1, Hoi Yin Loi2, Sophie O'Doherty3, Teng Hwee Tan4, Anthonin Reilhac3. 1. Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, 14 Medical Drive, #B1-01, Singapore, 117599, Singapore. mednyh@nus.edu.sg. 2. Department of Diagnostic Imaging, National University Hospital, Singapore, Singapore. 3. Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, 14 Medical Drive, #B1-01, Singapore, 117599, Singapore. 4. Department of Radiation Oncology, National University Cancer Institute, Singapore, Singapore.
Abstract
PURPOSE: To compare the performances of machine learning (ML) and deep learning (DL) in improving the quality of low dose (LD) lung cancer PET images and the minimum counts required. MATERIALS AND METHODS: 33 standard dose (SD) PET images, were used to simulate LD PET images at seven-count levels of 0.25, 0.5, 1, 2, 5, 7.5 and 10 million (M) counts. Image quality transfer (IQT), a ML algorithm that uses decision tree and patch-sampling was compared to two DL networks-HighResNet (HRN) and deep-boosted regression (DBR). Supervised training was performed by training the ML and DL algorithms with matched-pair SD and LD images. Image quality evaluation and clinical lesion detection tasks were performed by three readers. Bias in 53 radiomic features, including mean SUV, was evaluated for all lesions. RESULTS: ML- and DL-estimated images showed higher signal and smaller error than LD images with optimal image quality recovery achieved using LD down to 5 M counts. True positive rate and false discovery rate were fairly stable beyond 5 M counts for the detection of small and large true lesions. Readers rated average or higher ratings to images estimated from LD images of count levels above 5 M only, with higher confidence in detecting true lesions. CONCLUSION: LD images with a minimum of 5 M counts (8.72 MBq for 10 min scan or 25 MBq for 3 min scan) are required for optimal clinical use of ML and DL, with slightly better but more varied performance shown by DL.
PURPOSE: To compare the performances of machine learning (ML) and deep learning (DL) in improving the quality of low dose (LD) lung cancer PET images and the minimum counts required. MATERIALS AND METHODS: 33 standard dose (SD) PET images, were used to simulate LD PET images at seven-count levels of 0.25, 0.5, 1, 2, 5, 7.5 and 10 million (M) counts. Image quality transfer (IQT), a ML algorithm that uses decision tree and patch-sampling was compared to two DL networks-HighResNet (HRN) and deep-boosted regression (DBR). Supervised training was performed by training the ML and DL algorithms with matched-pair SD and LD images. Image quality evaluation and clinical lesion detection tasks were performed by three readers. Bias in 53 radiomic features, including mean SUV, was evaluated for all lesions. RESULTS: ML- and DL-estimated images showed higher signal and smaller error than LD images with optimal image quality recovery achieved using LD down to 5 M counts. True positive rate and false discovery rate were fairly stable beyond 5 M counts for the detection of small and large true lesions. Readers rated average or higher ratings to images estimated from LD images of count levels above 5 M only, with higher confidence in detecting true lesions. CONCLUSION: LD images with a minimum of 5 M counts (8.72 MBq for 10 min scan or 25 MBq for 3 min scan) are required for optimal clinical use of ML and DL, with slightly better but more varied performance shown by DL.
Authors: Eli Gibson; Wenqi Li; Carole Sudre; Lucas Fidon; Dzhoshkun I Shakir; Guotai Wang; Zach Eaton-Rosen; Robert Gray; Tom Doel; Yipeng Hu; Tom Whyntie; Parashkev Nachev; Marc Modat; Dean C Barratt; Sébastien Ourselin; M Jorge Cardoso; Tom Vercauteren Journal: Comput Methods Programs Biomed Date: 2018-01-31 Impact factor: 5.428