JTJustin Ruey Tse

Justin Ruey Tse

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Justin Ruey Tse has a total of 152 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Impact of VI-RADS quality score on VI-RADS diagnostic performance for muscle invasion: the role of tumor size and reader experience

    by Yohan Joo, Seong Woo Cho, Justin Ruey Tse, Jongjin Yoon, Dae Chul Jung, Byung Chul Kang, Young Taik Oh, Jae Hyon Park


  • Diagnostic Accuracy of Bosniak v2019 for Predicting Cancer and Unfavorable Histology in Class III–IV Cystic Renal Masses

    by Justin R. Tse, Shokoufeh Khalatbari, Matheos Yosef, Chen Abitbol, Asser M. Abou Elkassem, Emin Albayrak, Ronaldo Hueb Baroni, Rajesh Bhayana, Jennie J. Cao, Fernando Morbeck Almeida Coelho, et al.

    This multicenter multinational study demonstrated that Bosniak classification version 2019 can be used to stratify the probability of cancer and cancer with unfavorable histologic features; nodule size was strongly associated with unfavorable histologic features.


  • Site-specific limitations of ctDNA in PDAC surveillance: imaging detects pulmonary and local recurrence missed by liquid biopsy

    by Sherif B. Elsherif, Luyao Shen, Luke Yoon, Aya Kamaya, Justin R. Tse

    Objectives To evaluate the concordance between circulating tumor DNA (ctDNA) and imaging in the postoperative surveillance of patients with resected pancreatic ductal adenocarcinoma (PDAC). Methods This single-center, retrospective study included 50 patients with resected PDAC who underwent both ctDNA testing and surveillance imaging between January 2015 and February 2025. Imaging included CT or CT/MRI of the chest, abdomen, and pelvis at 3–6 month intervals, with ctDNA testing performed on a similar schedule. Two abdominal radiologists independently reviewed all imaging for recurrence, defined by new or enlarging lesions; discrepancies were resolved by consensus. ctDNA recurrence was defined as the detection of tumor-specific mutations. Concordance was assessed at the patient level. Results Among 50 patients (median age 69.5 years, 33 men, 17 women), ctDNA and imaging were concordant in 39 (78%): 22 had recurrence and 17 did not. Discordance occurred in 11 patients, 10 of whom had radiologic recurrence with negative ctDNA. Most discordant lesions were located in the lung (n = 5, median size 0.4 cm) or surgical bed (n = 5, median size 2.0 cm). Radiologic recurrences preceded negative ctDNA test results in all 10 of these cases. One patient had positive ctDNA without imaging recurrence over 497 days of follow-up. At the organ level, 1/6 (17%) of lung metastases and 10/15 (67%) of local recurrences by imaging were detected by ctDNA. All lymph node, liver, peritoneal, and bone metastases were detected by ctDNA. Conclusion Positive ctDNA tests after PDAC resection correlates with radiologic recurrences in the lymph nodes, liver, peritoneum, and bones. However, radiologic recurrences in the lungs or surgical bed are commonly associated with a negative ctDNA test. Radiologists should scrutinize the latter two sites even when ctDNA tests are negative.


  • Scout-Dose-TCM: direct and prospective scout-based estimation of personalized organ and effective doses from tube current modulated CT exams

    by Maria Jose Medrano, Sen Wang, Liyan Sun, Abdullah-Al-Zubaer Imran, Jennie Cao, Grant M Stevens, Justin R Tse, Adam S Wang

    Objective. This study proposes Scout-Dose-TCM for direct, prospective estimation of organ-level and effective doses under tube current modulation (TCM) and compares its performance with two established methods. Approach. Contrast-enhanced chest-abdomen-pelvis CT exams from 130 adults (120 kVp, TCM) were analyzed. Reference organ doses for six organs (lungs, kidneys, liver, pancreas, bladder, spleen) were calculated using Monte Carlo (MC)-GPU and TotalSegmentator. These data trained Scout-Dose-TCM, a deep-learning model that predicts organ-level doses corresponding to discrete cosine transform (DCT) basis functions, enabling real-time estimation for any TCM profile. The model includes a feature-learning module that extracts contextual information from lateral and frontal scouts and scan range, and a dose-learning module that outputs DCT-based dose estimates. The loss function incorporated the DCT formulation, ensuring accurate predictions across variable TCM patterns. For comparison, dose estimation was performed following AAPM Task Group 204 (Global CTDI vol ) and its TCM-adapted and organ-specific version (Organ CTDI vol ). The three evaluated methods were extended to estimate effective dose and compared against the approach in AAPM Report 96, which applies a fixed dose-length product-to-effective-dose conversion factor. Five-fold cross-validation assessed generalizability via mean absolute percentage dose errors and R 2 correlations with MC benchmarks. Main Results. Mean organ-level absolute percentage errors were 13% (Global CTDI vol ), 9% (Organ CTDI vol ), and 7% (Scout-Dose-TCM). The largest discrepancies occurred for the bladder (15%, 13%, and 9%). Scout-Dose-TCM significantly reduced organ-level dose errors versus Global CTDI vol ( p < 1 × 10 −7 ) and improved predictions for liver, bladder, and pancreas versus Organ CTDI vol ( p ⩽ 0.005). It also achieved higher R 2 values and lower effective-dose error (3.01%) than Global (3.87%) and Organ (4.16%) CTDI vol , and the AAPM Task Group 96 method (31%). Significance. Scout-Dose-TCM outperformed Global CTDI vol and was comparable to or better than Organ CTDI vol for organ and effective dose estimation without requiring segmentations at inference, demonstrating its clinical potential for prospective dose estimation in CT.


  • Imaging surveillance outcomes of hepatocellular adenomas: systematic review and meta-analysis

    by Luyao Shen, Stephan Altmayer, Justin Ruey Tse

    Objectives Hepatocellular adenomas (HCA) are typically resected if > 5 cm after 6 months due to historical risks of bleeding and malignant transformation. Outcomes of contemporary patients managed with surveillance are not well defined. We performed a systematic review and meta-analysis of patients with HCAs monitored for ≥ 6 months to assess rates of size stability, bleeding, and malignant transformation. Methods PubMed, Web of Science, and Embase databases were searched up to August 2025 for all articles reporting HCA and follow-up. Risk of bias was assessed with modified Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool. Pooled proportions of outcomes were assessed with linear mixed model and logistic regression. Heterogeneity was assessed with the I2 index. Results Of 679 articles screened, 7 retrospective studies (2019–2023) with 699 patients across Europe and United States were included. Five were multicenter. Median follow-up ranged from 2 to 5.8 years. For size stability, four studies used Response Evaluation Criteria in Solid Tumors (RECIST); others used a 1 cm size threshold, regression to < 5 cm, or unspecified criteria. Six studies showed moderate to serious risk of bias. Pooled proportion of HCAs that were stable or regressed was 88.5%. Pooled incidences of symptomatic bleeding and malignant transformation were 0.5% and 0.8%, respectively. Conclusion In a contemporary series, the majority of untreated HCAs were stable or regressed during surveillance, with extremely low incidences (< 1%) of bleeding and malignant transformation. Extending surveillance intervals may safely offer patients more time to address modifiable risk factors.


  • The Current Role of Renal Mass Biopsy

    by Luyao Shen, Daniel H. Kwon, Geoffrey A. Sonn, Justin R. Tse


  • Automated estimation of individualized organ-specific dose and noise from clinical CT scans

    by Sen Wang, Maria Jose Medrano, Abdullah Al Zubaer Imran, Wonkyeong Lee, Jennie Jiayi Cao, Grant M Stevens, Justin Ruey Tse, Adam S Wang

    Objective. Radiation dose and diagnostic image quality are opposing constraints in x-ray computed tomography (CT). Conventional methods do not fully account for organ-level radiation dose and noise when considering radiation risk and clinical task. In this work, we develop a pipeline to generate individualized organ-specific dose and noise at desired dose levels from clinical CT scans. Approach. To estimate organ-specific dose and noise, we compute dose maps, noise maps at desired dose levels and organ segmentations. In our pipeline, dose maps are generated using Monte Carlo simulation. The noise map is obtained by scaling the inserted noise in synthetic low-dose emulation in order to avoid anatomical structures, where the scaling coefficients are empirically calibrated. Organ segmentations are generated by a deep learning-based method (TotalSegmentator). The proposed noise model is evaluated on a clinical dataset of 12 CT scans, a phantom dataset of 3 uniform phantom scans, and a cross-site dataset of 26 scans. The accuracy of deep learning-based segmentations for organ-level dose and noise estimates was tested using a dataset of 41 cases with expert segmentations of six organs: lungs, liver, kidneys, bladder, spleen, and pancreas. Main results. The empirical noise model performs well, with an average RMSE approximately 1.5 HU and an average relative RMSE approximately 5% across different dose levels. The segmentation from TotalSegmentator yielded a mean Dice score of 0.8597 across the six organs (max = 0.9315 in liver, min = 0.6855 in pancreas). The resulting error in organ-level dose and noise estimation was less than 2% for most organs. Significance. The proposed pipeline can output individualized organ-specific dose and noise estimates accurately for personalized protocol evaluation and optimization. It is fully automated and can be scalable to large clinical datasets. This pipeline can be used to optimize image quality for specific organs and thus clinical tasks, without adversely affecting overall radiation dose.


  • Varied terminology by radiologists to describe Bosniak class III and IV cystic renal masses

    by Acacia Yoon, Justin R Tse

    Background: The Bosniak classification is designed to standardize evaluation of cystic renal masses and to communicate the risk of malignancy. Purpose: To determine whether radiologists vary in their communication of Bosniak class III and IV cystic renal masses. Material and Methods: This retrospective study included 186 patients with CT or MRI reporting a Bosniak class III or IV mass. Radiology reports were evaluated to determine the noun representing the mass, the modifier to convey the likelihood of cancer, and recommendations for urologic referral. Electronic medical records were reviewed to determine if the patient saw a urologist. Results: Of the patients, 112 (60%) had a class III mass and 74 (40%) had a class IV mass. Class III masses were more likely to be represented by the noun “lesion” rather than a “mass” (61/112 [54%] vs. 31/112 [28%]). Class IV masses are more likely to be represented as a “mass” (36/74 [59%] vs. 28/74 [38%]; P < 0.015). Cancer was described in 100/186 (54%) cases: 38/112 (35%) class III masses and 62/74 (72%) class IV masses ( P < 0.001). The cancer terminology used included “renal cell carcinoma” (n = 57), “neoplasm” (n = 12), and “malignancy/malignant” (n = 86). Most radiology reports (n = 133, 72%) did not recommend urologic referral but 183 (98%) patients were referred and 181 (97%) ultimately saw a urologist. Conclusion: Radiologists vary in their communication of class III and IV masses, reflecting historical terminology usage, nuanced interpretations, and an evolving understanding of renal cell carcinoma biology. This variance had minimal impact on urologic referral rates.