Does image resolution impact chest X-ray based fine-grained Tuberculosis-consistent lesion segmentation?
This work addresses a practical problem for medical imaging researchers by optimizing computational efficiency in TB lesion segmentation, though it is incremental as it focuses on resolution tuning rather than novel methodology.
The study investigated the impact of image resolution on fine-grained Tuberculosis-consistent lesion segmentation in chest X-rays, finding that higher resolutions are not always necessary but identifying the optimal resolution is critical for superior performance, with specific improvements using a combinatorial approach.
Deep learning (DL) models are state-of-the-art in segmenting anatomical and disease regions of interest (ROIs) in medical images. Particularly, a large number of DL-based techniques have been reported using chest X-rays (CXRs). However, these models are reportedly trained on reduced image resolutions for reasons related to the lack of computational resources. Literature is sparse in discussing the optimal image resolution to train these models for segmenting the Tuberculosis (TB)-consistent lesions in CXRs. In this study, we investigated the performance variations using an Inception-V3 UNet model using various image resolutions with/without lung ROI cropping and aspect ratio adjustments, and (ii) identified the optimal image resolution through extensive empirical evaluations to improve TB-consistent lesion segmentation performance. We used the Shenzhen CXR dataset for the study which includes 326 normal patients and 336 TB patients. We proposed a combinatorial approach consisting of storing model snapshots, optimizing segmentation threshold and test-time augmentation (TTA), and averaging the snapshot predictions, to further improve performance with the optimal resolution. Our experimental results demonstrate that higher image resolutions are not always necessary, however, identifying the optimal image resolution is critical to achieving superior performance.