Weakly supervised 3D classification of chest CT using aggregated multi-resolution deep segmentation features

Abstract

Weakly supervised disease classification of CT imaging suffers from poor localization owing to case-level annotations, where even a positive scan can hold hundreds to thousands of negative slices along multiple planes. Furthermore, although deep learning segmentation and classification models extract distinctly unique combinations of anatomical features from the same target class(es), they are typically seen as two independent processes in a computer-aided diagnosis (CAD) pipeline, with little to no feature reuse. In this research, we propose a medical classifier that leverages the semantic structural concepts learned via multi-resolution segmentation feature maps, to guide weakly supervised 3D classification of chest CT volumes. Additionally, a comparative analysis is drawn across two different types of feature aggregation to explore the vast possibilities surrounding feature fusion. Using a dataset of 1593 scans labeled on a case-level basis via rule-based model, we train a dual-stage convolutional neural network (CNN) to perform organ segmentation and binary classification of four representative diseases (emphysema, pneumonia/atelectasis, mass and nodules) in lungs. The baseline model, with separate stages for segmentation and classification, results in AUC of 0.791. Using identical hyperparameters, the connected architecture using static and dynamic feature aggregation improves performance to AUC of 0.832 and 0.851, respectively. This study advances the field in two key ways. First, case-level report data is used to weakly supervise a 3D CT classifier of multiple, simultaneous diseases for an organ. Second, segmentation and classification models are connected with two different feature aggregation strategies to enhance the classification performance.

Reinforcement Learning Driven Intra-modal and Inter-modal Representation Learning for 3D Medical Image Classification

“…In addition, there are several works that applied DL networks for multi-modal learning [20,11,15,13,9]. These methods can be roughly categorised into two branches: 2D-based methods [19,10] and 3D-based methods [5,8,14]. For the first line of methods, 3D volumes are firstly projected into 2D images and then are integrated for the final prediction.…”

Tailored 3D CT contrastive pretraining to improve pulmonary pathology classification

“…There has been substantial prior work on multipleabnormality prediction in different modalities; chest CT [1], body CT [2], chest x-ray for COVID-19 [3] and 3D MRI for brain tumor classification [4]. However, constructing multiclass (i.e., pathology) classifiers for 3D medical data isn't a straightforward undertaking and remains challenging because of two main obstacles: acquiring sufficiently large datasets and learning effective representations with weak pathological footprints.…”