arXiv subjects

Justin R. Tse

Publications and source records attributed to Justin R. Tse.

2 recordsLinked to original sources

arXiv2026

Liver Metastasis Detection at Reduced Radiation Dose: Diagnostic Evaluation of a Novel Organ-Level Tube Current Modulation Method

Objective: To evaluate retroOpt, a novel organ-level tube current modulation (TCM) method that minimizes effective radiation dose while preserving diagnostic image quality for liver metastasis detection. Methods: In this retrospective, IRB-approved study, 22 patients with 68 liver lesions (38 malignant, 30 benign) underwent portal venous phase contrast-enhanced chest-abdomen-pelvis CT. Using projection-domain noise emulation, five series were generated per patient: original full dose (Orig-100), uniform dose reduction to 40% (UD-40) and 60% (UD-60) of the original effective dose, and organ-level TCM at the same levels (Opt-40, Opt-60). Three abdominal radiologists independently detected and classified lesions and rated image quality (5-point Likert). Per-lesion sensitivity was compared by McNemar test. Malignant lesion-size thresholds for 50% and 90% sensitivity (x50, x90) were estimated by logistic regression. Image quality was compared by Wilcoxon signed-rank test. Results: At the 60% level, Opt-60 achieved mean malignant lesion sensitivity comparable to full-dose CT (82% vs 80%) and exceeded uniform reduction (67%). At the 40% level, Opt-40 improved mean sensitivity over UD-40 from 52% to 68%. Mean all-lesion sensitivity rose from 54% (UD-40) to 64% (Opt-40) and from 65% (UD-60) to 77% (Opt-60), remaining comparable to full dose (76%). All optimized-versus-uniform differences were significant (p <= 0.003). Opt-60 lesion-size thresholds closely matched Orig-100 (x50 3.3 vs 3.7 mm; x90 15.7 vs 16.8 mm). Median Opt-60 image quality ranged from 3 to 4.5 across readers. Conclusions: Organ-level TCM preserved liver metastasis detection at 60% of the original effective dose, outperforming uniform dose reduction. Task-specific organ-level dose optimization may enable greater CT dose reduction than uniform strategies for patients requiring repeated metastasis surveillance.

physics.med-ph↗

arXiv2022

Correlated Feature Aggregation by Region Helps Distinguish Aggressive from Indolent Clear Cell Renal Cell Carcinoma Subtypes on CT

Renal cell carcinoma (RCC) is a common cancer that varies in clinical behavior. Indolent RCC is often low-grade without necrosis and can be monitored without treatment. Aggressive RCC is often high-grade and can cause metastasis and death if not promptly detected and treated. While most kidney cancers are detected on CT scans, grading is based on histology from invasive biopsy or surgery. Determining aggressiveness on CT images is clinically important as it facilitates risk stratification and treatment planning. This study aims to use machine learning methods to identify radiology features that correlate with features on pathology to facilitate assessment of cancer aggressiveness on CT images instead of histology. This paper presents a novel automated method, Correlated Feature Aggregation By Region (CorrFABR), for classifying aggressiveness of clear cell RCC by leveraging correlations between radiology and corresponding unaligned pathology images. CorrFABR consists of three main steps: (1) Feature Aggregation where region-level features are extracted from radiology and pathology images, (2) Fusion where radiology features correlated with pathology features are learned on a region level, and (3) Prediction where the learned correlated features are used to distinguish aggressive from indolent clear cell RCC using CT alone as input. Thus, during training, CorrFABR learns from both radiology and pathology images, but during inference, CorrFABR will distinguish aggressive from indolent clear cell RCC using CT alone, in the absence of pathology images. CorrFABR improved classification performance over radiology features alone, with an increase in binary classification F1-score from 0.68 (0.04) to 0.73 (0.03). This demonstrates the potential of incorporating pathology disease characteristics for improved classification of aggressiveness of clear cell RCC on CT images.

eess.IV↗