Hye Joo Son1, Jungsu S Oh1, Minyoung Oh1, Soo Jong Kim1, Jae-Hong Lee2, Jee Hoon Roh3, Jae Seung Kim4. 1. Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea. 2. Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea. 3. Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea. alzheimer@naver.com. 4. Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea. jaeskim@amc.seoul.kr.
Abstract
PURPOSE: Although most deep learning (DL) studies have reported excellent classification accuracy, these studies usually target typical Alzheimer's disease (AD) and normal cognition (NC) for which conventional visual assessment performs well. A clinically relevant issue is the selection of high-risk subjects who need active surveillance among equivocal cases. We validated the clinical feasibility of DL compared with visual rating or quantitative measurement for assessing the diagnosis and prognosis of subjects with equivocal amyloid scans. METHODS: 18F-florbetaben scans of 430 cases (85 NC, 233 mild cognitive impairment, and 112 AD) were assessed through visual rating-based, quantification-based, and DL-based methods. DL was trained using 280 two-dimensional PET images (80%) and tested by randomly assigning the remaining (70 cases, 20%) cases and a clinical validation set of 54 equivocal cases. In the equivocal cases, we assessed the agreement among the visual rating, quantification, and DL and compared the clinical outcome according to each modality-based amyloid status. RESULTS: The visual reading was positive in 175 cases, equivocal in 54 cases, and negative in 201 cases. The composite SUVR cutoff value was 1.32 (AUC 0.99). The subject-level performance of DL using the test set was 100%. Among the 54 equivocal cases, 37 cases were classified as positive (Eq(deep+)) by DL, 40 cases were classified by a second-round visual assessment, and 40 cases were classified by quantification. The DL- and quantification-based classifications showed good agreement (83%, κ = 0.59). The composite SUVRs differed between Eq(deep+) (1.47 [0.13]) and Eq(deep-) (1.29 [0.10]; P < 0.001). DL, but not the visual rating, showed a significant difference in the Mini-Mental Status Examination score change during the follow-up between Eq(deep+) (- 4.21 [0.57]) and Eq(deep-) (- 1.74 [0.76]; P = 0.023) (mean duration, 1.76 years). CONCLUSIONS: In visually equivocal scans, DL was more related to quantification than to visual assessment, and the negative cases selected by DL showed no decline in cognitive outcome. DL is useful for clinical diagnosis and prognosis assessment in subjects with visually equivocal amyloid scans.
PURPOSE: Although most deep learning (DL) studies have reported excellent classification accuracy, these studies usually target typical Alzheimer's disease (AD) and normal cognition (NC) for which conventional visual assessment performs well. A clinically relevant issue is the selection of high-risk subjects who need active surveillance among equivocal cases. We validated the clinical feasibility of DL compared with visual rating or quantitative measurement for assessing the diagnosis and prognosis of subjects with equivocal amyloid scans. METHODS: 18F-florbetaben scans of 430 cases (85 NC, 233 mild cognitive impairment, and 112 AD) were assessed through visual rating-based, quantification-based, and DL-based methods. DL was trained using 280 two-dimensional PET images (80%) and tested by randomly assigning the remaining (70 cases, 20%) cases and a clinical validation set of 54 equivocal cases. In the equivocal cases, we assessed the agreement among the visual rating, quantification, and DL and compared the clinical outcome according to each modality-based amyloid status. RESULTS: The visual reading was positive in 175 cases, equivocal in 54 cases, and negative in 201 cases. The composite SUVR cutoff value was 1.32 (AUC 0.99). The subject-level performance of DL using the test set was 100%. Among the 54 equivocal cases, 37 cases were classified as positive (Eq(deep+)) by DL, 40 cases were classified by a second-round visual assessment, and 40 cases were classified by quantification. The DL- and quantification-based classifications showed good agreement (83%, κ = 0.59). The composite SUVRs differed between Eq(deep+) (1.47 [0.13]) and Eq(deep-) (1.29 [0.10]; P < 0.001). DL, but not the visual rating, showed a significant difference in the Mini-Mental Status Examination score change during the follow-up between Eq(deep+) (- 4.21 [0.57]) and Eq(deep-) (- 1.74 [0.76]; P = 0.023) (mean duration, 1.76 years). CONCLUSIONS: In visually equivocal scans, DL was more related to quantification than to visual assessment, and the negative cases selected by DL showed no decline in cognitive outcome. DL is useful for clinical diagnosis and prognosis assessment in subjects with visually equivocal amyloid scans.
Entities:
Keywords:
18F-florbetaben PET; Alzheimer’s disease; Amyloid; Deep learning; Equivocal scan
Authors: Bruce W Patterson; Donald L Elbert; Kwasi G Mawuenyega; Tom Kasten; Vitaliy Ovod; Shengmei Ma; Chengjie Xiong; Robert Chott; Kevin Yarasheski; Wendy Sigurdson; Lily Zhang; Alison Goate; Tammie Benzinger; John C Morris; David Holtzman; Randall J Bateman Journal: Ann Neurol Date: 2015-07-20 Impact factor: 10.422
Authors: Gengyan Zhao; Fang Liu; Jonathan A Oler; Mary E Meyerand; Ned H Kalin; Rasmus M Birn Journal: Neuroimage Date: 2018-03-28 Impact factor: 6.556
Authors: Renaud La Joie; Nagehan Ayakta; William W Seeley; Ewa Borys; Adam L Boxer; Charles DeCarli; Vincent Doré; Lea T Grinberg; Eric Huang; Ji-Hye Hwang; Milos D Ikonomovic; Clifford Jack; William J Jagust; Lee-Way Jin; William E Klunk; Julia Kofler; Orit H Lesman-Segev; Samuel N Lockhart; Val J Lowe; Colin L Masters; Chester A Mathis; Catriona L McLean; Bruce L Miller; Daniel Mungas; James P O'Neil; John M Olichney; Joseph E Parisi; Ronald C Petersen; Howard J Rosen; Christopher C Rowe; Salvatore Spina; Prashanthi Vemuri; Victor L Villemagne; Melissa E Murray; Gil D Rabinovici Journal: Alzheimers Dement Date: 2018-10-19 Impact factor: 21.566
Authors: Yiming Ding; Jae Ho Sohn; Michael G Kawczynski; Hari Trivedi; Roy Harnish; Nathaniel W Jenkins; Dmytro Lituiev; Timothy P Copeland; Mariam S Aboian; Carina Mari Aparici; Spencer C Behr; Robert R Flavell; Shih-Ying Huang; Kelly A Zalocusky; Lorenzo Nardo; Youngho Seo; Randall A Hawkins; Miguel Hernandez Pampaloni; Dexter Hadley; Benjamin L Franc Journal: Radiology Date: 2018-11-06 Impact factor: 29.146
Authors: Rik Vandenberghe; Natalie Nelissen; Eric Salmon; Adrian Ivanoiu; Steen Hasselbalch; Allan Andersen; Alex Korner; Lennart Minthon; David J Brooks; Koen Van Laere; Patrick Dupont Journal: Neuroimage Date: 2012-09-14 Impact factor: 6.556
Authors: Dimitris Visvikis; Philippe Lambin; Kim Beuschau Mauridsen; Roland Hustinx; Michael Lassmann; Christoph Rischpler; Kuangyu Shi; Jan Pruim Journal: Eur J Nucl Med Mol Imaging Date: 2022-07-09 Impact factor: 9.236
Authors: Ying-Hwey Nai; Yee-Hsin Tay; Tomotaka Tanaka; Christopher P Chen; Edward G Robins; Anthonin Reilhac Journal: Neuroinformatics Date: 2022-05-27
Authors: James Zou; David Park; Aubrey Johnson; Xinyang Feng; Michelle Pardo; Jeanelle France; Zeljko Tomljanovic; Adam M Brickman; Devangere P Devanand; José A Luchsinger; William C Kreisl; Frank A Provenzano Journal: Alzheimers Dement (Amst) Date: 2021-12-31
Authors: Ashwin V Venkataraman; Wenjia Bai; Alex Whittington; James F Myers; Eugenii A Rabiner; Anne Lingford-Hughes; Paul M Matthews Journal: Alzheimers Res Ther Date: 2021-11-10 Impact factor: 6.982