Weakly-supervised learning-based pathology detection and localization in 3D chest CT scans.
3D pathology localization
computed tomography
multi‐abnormality detection
self‐supervised learning
weakly‐supervised learning
Journal
Medical physics
ISSN: 2473-4209
Titre abrégé: Med Phys
Pays: United States
ID NLM: 0425746
Informations de publication
Date de publication:
14 Aug 2024
14 Aug 2024
Historique:
revised:
24
05
2024
received:
11
12
2023
accepted:
01
07
2024
medline:
14
8
2024
pubmed:
14
8
2024
entrez:
14
8
2024
Statut:
aheadofprint
Résumé
Recent advancements in anomaly detection have paved the way for novel radiological reading assistance tools that support the identification of findings, aimed at saving time. The clinical adoption of such applications requires a low rate of false positives while maintaining high sensitivity. In light of recent interest and development in multi pathology identification, we present a novel method, based on a recent contrastive self-supervised approach, for multiple chest-related abnormality identification including low lung density area ("LLDA"), consolidation ("CONS"), nodules ("NOD") and interstitial pattern ("IP"). Our approach alerts radiologists about abnormal regions within a computed tomography (CT) scan by providing 3D localization. We introduce a new method for the classification and localization of multiple chest pathologies in 3D Chest CT scans. Our goal is to distinguish four common chest-related abnormalities: "LLDA", "CONS", "NOD", "IP" and "NORMAL". This method is based on a 3D patch-based classifier with a Resnet backbone encoder pretrained leveraging recent contrastive self supervised approach and a fine-tuned classification head. We leverage the SimCLR contrastive framework for pretraining on an unannotated dataset of randomly selected patches and we then fine-tune it on a labeled dataset. During inference, this classifier generates probability maps for each abnormality across the CT volume, which are aggregated to produce a multi-label patient-level prediction. We compare different training strategies, including random initialization, ImageNet weight initialization, frozen SimCLR pretrained weights and fine-tuned SimCLR pretrained weights. Each training strategy is evaluated on a validation set for hyperparameter selection and tested on a test set. Additionally, we explore the fine-tuned SimCLR pretrained classifier for 3D pathology localization and conduct qualitative evaluation. Validated on 111 chest scans for hyperparameter selection and subsequently tested on 251 chest scans with multi-abnormalities, our method achieves an AUROC of 0.931 (95% confidence interval [CI]: [0.9034, 0.9557], The proposed method integrates self-supervised learning algorithms for pretraining, utilizes a patch-based approach for 3D pathology localization and develops an aggregation method for multi-label prediction at patient-level. It shows promise in efficiently detecting and localizing multiple anomalies within a single scan.
Sections du résumé
BACKGROUND
BACKGROUND
Recent advancements in anomaly detection have paved the way for novel radiological reading assistance tools that support the identification of findings, aimed at saving time. The clinical adoption of such applications requires a low rate of false positives while maintaining high sensitivity.
PURPOSE
OBJECTIVE
In light of recent interest and development in multi pathology identification, we present a novel method, based on a recent contrastive self-supervised approach, for multiple chest-related abnormality identification including low lung density area ("LLDA"), consolidation ("CONS"), nodules ("NOD") and interstitial pattern ("IP"). Our approach alerts radiologists about abnormal regions within a computed tomography (CT) scan by providing 3D localization.
METHODS
METHODS
We introduce a new method for the classification and localization of multiple chest pathologies in 3D Chest CT scans. Our goal is to distinguish four common chest-related abnormalities: "LLDA", "CONS", "NOD", "IP" and "NORMAL". This method is based on a 3D patch-based classifier with a Resnet backbone encoder pretrained leveraging recent contrastive self supervised approach and a fine-tuned classification head. We leverage the SimCLR contrastive framework for pretraining on an unannotated dataset of randomly selected patches and we then fine-tune it on a labeled dataset. During inference, this classifier generates probability maps for each abnormality across the CT volume, which are aggregated to produce a multi-label patient-level prediction. We compare different training strategies, including random initialization, ImageNet weight initialization, frozen SimCLR pretrained weights and fine-tuned SimCLR pretrained weights. Each training strategy is evaluated on a validation set for hyperparameter selection and tested on a test set. Additionally, we explore the fine-tuned SimCLR pretrained classifier for 3D pathology localization and conduct qualitative evaluation.
RESULTS
RESULTS
Validated on 111 chest scans for hyperparameter selection and subsequently tested on 251 chest scans with multi-abnormalities, our method achieves an AUROC of 0.931 (95% confidence interval [CI]: [0.9034, 0.9557],
CONCLUSIONS
CONCLUSIONS
The proposed method integrates self-supervised learning algorithms for pretraining, utilizes a patch-based approach for 3D pathology localization and develops an aggregation method for multi-label prediction at patient-level. It shows promise in efficiently detecting and localizing multiple anomalies within a single scan.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Informations de copyright
© 2024 American Association of Physicists in Medicine.
Références
Ardila D, Kiraly AP, Bharadwaj S, et al. End‐to‐end lung cancer screening with three‐dimensional deep learning on low‐dose chest computed tomography. Nat Med. 2019;25:954‐961.
Wang S, Dong L, Wang X, Wang X. Classification of pathological types of lung cancer from CT images by deep residual neural networks with transfer learning strategy. Open Med (Wars). 2020;15:190‐197.
Liu G, Liu F, Gu J, Mao X, Xie X, Sang J. An attention‐based deep learning network for lung nodule malignancy discrimination. Front Neurosci. 2022;16:1106937.
Liu D, Liu F, Tie Y, Qi L, Wang F. Res‐trans networks for lung nodule classification. Int J Comput Assist Radiol Surg. 2022;17:1059‐1068.
Salama WM, Shokry A, Aly MH. A generalized framework for lung cancer classification based on deep generative models. Multimed Tools Appl. 2022;81:32705‐32722.
Bai Y, Li D, Duan Q, Chen X. Analysis of high‐resolution reconstruction of medical images based on deep convolutional neural networks in lung cancer diagnostics. Comput Methods Programs Biomed. 2022;217:106592.
Bermejo‐Peláez D, Ash SY, Washko GR, San José Estépar R, Ledesma‐Carbayo MJ. Classification of interstitial lung abnormality patterns with an ensemble of deep convolutional neural networks. Sci Rep. 2020;10:338.
Gao M, Bagci U, Lu L, et al. Holistic classification of CT attenuation patterns for interstitial lung diseases via deep convolutional neural networks. Comput Methods Biomech Biomed Eng Imaging Vis. 2018;6:1‐6.
Anthimopoulos M, Christodoulidis S, Ebner L, Christe A, Mougiakakou S. Lung pattern classification for interstitial lung diseases using a deep convolutional neural network. IEEE Trans Med Imaging. 2016;35:1207‐1216.
Gao M, Xu Z, Lu L, Harrison A, Summers R, Mollura D. Multi‐label Deep Regression and Unordered Pooling for Holistic Interstitial Lung Disease Detection. Springer International Publishing; 2016.
Wang C, Moriya T, Hayashi Y, et al. Weakly‐supervised deep learning of interstitial lung disease types on CT images. In: Mori K, Hahn HK, eds. Medical Imaging 2019: Computer‐Aided Diagnosis. International Society for Optics and Photonics. SPIE; 2019:109501H. doi:10.1117/12.2512746
Christodoulidis S, Anthimopoulos M, Ebner L, Christe A, Mougiakakou S. Multisource transfer learning with convolutional neural networks for lung pattern analysis. IEEE J Biomed Health Inform. 2017;21:76‐84.
Chaddad A, Hassan L, Desrosiers C. Deep CNN models for predicting COVID‐19 in CT and x‐ray images. J Med Imaging. 2021;8:014502.
Rana A, Singh H, Mavuduru R, Pattanaik S, Rana PS. Quantifying prognosis severity of COVID‐19 patients from deep learning based analysis of CT chest images. Multimed Tools Appl. 2022;81:18129‐18153.
Shah V, Keniya R, Shridharani A, Punjabi M, Shah J, Mehendale N. Diagnosis of COVID‐19 using CT scan images and deep learning techniques. Emerg Radiol. 2021;28:497‐505.
Yang D, Martinez C, Visuña L, Khandhar H, Bhatt C, Carretero J. Detection and analysis of COVID‐19 in medical images using deep learning techniques. Sci Rep. 2021;11:19638.
Zhao W, Jiang W, Qiu X. Deep learning for COVID‐19 detection based on CT images. Sci Rep. 2021;11:14353.
Fuhrman J, Yip R, Zhu Y, et al. Evaluation of emphysema on thoracic low‐dose CTs through attention‐based multiple instance deep learning. Sci Rep. 2023;13:1187.
Nagaraj Y, Wisselink HJ, Rook M, et al. AI‐driven model for automatic emphysema detection in low‐dose computed tomography using disease‐specific augmentation. J Digit Imaging. 2022;35:538‐550.
Draelos RL, Dov D, Mazurowski MA, et al. Machine‐learning‐based multiple abnormality prediction with large‐scale chest computed tomography volumes. Med Image Anal. 2021;67:101857. doi:10.1016/j.media.2020.101857
Tushar FI, D'Anniballe VM, Hou R, et al. Classification of multiple diseases on body CT scans using weakly supervised deep learning. Radiol Artif Intell. 2022;4:e210026.
Saha A, Tushar FI, Faryna K, et al. Weakly supervised 3D classification of chest CT using aggregated multi‐resolution deep segmentation features. In: Medical Imaging 2020: Computer‐Aided Diagnosis. SPIE; 2020:1131408.
Chen T, Kornblith S, Norouzi M, Hinton G. A simple framework for contrastive learning of visual representations. In: Proceedings of the 37th International Conference on Machine Learning. PMLR; 2020:149. doi:10.5555/3524938.3525087
Genin G, Craighero F. Chapter 2. In: Scanner thoracique, guide de lecture rapide: images clefs, définitions et étiologies. Sauramps Medical. Librairie Dalloz; 2020. Available at: https://www.librairiedalloz.fr/livre/9791030302714‐scanner‐thoracique‐guide‐de‐lecture‐rapide‐images‐clefs‐definitions‐et‐etiologies‐gilles‐genin‐fabien‐craighero/
Djahnine A, Popoff A, Jupin‐Delevaux E, Cotin V, Nempont O, Boussel L. Tailored 3D CT contrastive pretraining to improve pulmonary pathology classification. In: 2022 16th IEEE International Conference on Signal Processing (ICSP). IEEE; 2022:229‐234. doi:10.1109/ICSP56322.2022.9965218
Si‐Mohamed SA, Nasser M, Colevray M, et al. Automatic quantitative computed tomography measurement of longitudinal lung volume loss in interstitial lung diseases. Eur Radiol. 2022;32(6):4292‐4303. doi:10.1007/s00330‐021‐08482‐9
Galzin E, Roche L, Vlachomitrou A, et al. Additional value of chest CT AI‐based quantification of lung involvement in predicting death and ICU admission for COVID‐19 patients. Res Diagn Interv Imaging. 2022;4:100018. doi:10.1016/j.redii.2022.100018