Biomedical Pattern Recognition
-
Indifference Subspace of Deep Features for Lung Nodule Classification from CT Images
This work examines whether standard DL networks can produce common feature node magnitudes for same-class images. Surprisingly, the indifference subspace that emerges from standard DL networks performs remarkably, allowing fine-tuning with CVA and yielding classification performances on par with state-of-the-art… Continue reading
-
Are Deep Learning Classification Results Obtained on CT Scans Fair and Interpretable?
The unfair model often incorrectly focused on non-tumor areas for malignancy predictions, while the fair model correctly concentrated on tumor regions. This indicates the unfair model’s unreliability, likely stemming from overfitting due to improper train/test splitting. Continue reading
-
Deep Learning Enhancement of Low-Dose CT Images for Improved Diagnostic Accuracy
The use of ionizing radiation in diagnostic imaging is a common practice worldwide. However, the imaging process itself carries a relative risk. Therefore, it is recommended to employ the lowest possible dose of ionizing radiation, especially in computed tomography (CT)… Continue reading
Pattern Recognition Research Lab. 