Steve Y. Cho
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
Research Subjects
Steve Y. Cho has a total of 170 co-authors. Here are the top co-authors by number of shared publications:
Steven P. Rowe
13 shared publicationsKenneth L. Gage
10 shared publications- Andrei Iagaru
Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, Stanford University, Stanford, CA
9 shared publications Austin R. Pantel
9 shared publications- Barry A. Siegel
Mallinckrodt Institute of Radiology, Washington University, St. Louis, MO, USA
9 shared publications David Josephson
9 shared publications
Latest Publications
- 1 May 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (3) , e240229
Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Using a Longitudinally Aware Segmentation Network
A longitudinally aware segmentation network trained on multicenter clinical trial data achieved high performance in automatic quantification of PET metrics for baseline and interim-therapy scans in pediatric Hodgkin lymphoma.
- 1 May 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (3) , e240229
Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Using a Longitudinally Aware Segmentation Network
A longitudinally aware segmentation network trained on multicenter clinical trial data achieved high performance in automatic quantification of PET metrics for baseline and interim-therapy scans in pediatric Hodgkin lymphoma.
- 1 January 2025
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 314 (1) , e232650
CT, MRI, and FDG PET/CT in the Assessment of Lymph Node Involvement in Pediatric Hodgkin Lymphoma: An Expert Consensus Definition by an International Collaboration on Staging Evaluation and Response Criteria Harmonization for Children, Adolescent, and Young Adult Hodgkin Lymphoma (SEARCH for CAYAHL)
An international consensus definition of lymph node involvement at CT, MRI, and fluorodeoxyglucose PET/CT in pediatric Hodgkin lymphoma allows more robust comparisons between clinical trials, particularly those that include involved node or involved site radiation therapy.
- 1 January 2025
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 314 (1) , e232650
CT, MRI, and FDG PET/CT in the Assessment of Lymph Node Involvement in Pediatric Hodgkin Lymphoma: An Expert Consensus Definition by an International Collaboration on Staging Evaluation and Response Criteria Harmonization for Children, Adolescent, and Young Adult Hodgkin Lymphoma (SEARCH for CAYAHL)
An international consensus definition of lymph node involvement at CT, MRI, and fluorodeoxyglucose PET/CT in pediatric Hodgkin lymphoma allows more robust comparisons between clinical trials, particularly those that include involved node or involved site radiation therapy.
- Open Access
- 30 December 2024
- Published by Wolters Kluwer Health in Clinical Nuclear Medicine Open
- Vol. 1 (2)
Interobserver Reliability of the 5-Point Deauville Score and SUV-Based Quantification of FDG PET/CT Scans in High-Risk Pediatric Hodgkin Lymphoma
Purpose: In lymphomas, the Deauville score (DS) standardizes 18F-FDG PET treatment response assessments. We assessed interobserver agreement of the DS and developed a quantitative PET-based parameter to minimize subjectivity of response assessments. Methods: PET scans performed within the Children’s Oncology Group AHOD0831 study were analyzed. One hundred one scans obtained after the first cycle of chemotherapy (PET1) and 83 obtained after the second cycle (PET2) were eligible for inclusion. Each scan was assigned a DS by 2 nuclear medicine radiologists and a Hodgkin lymphoma index (HLI), defined as the ratio of the tumor’s SUVmax or SUVpeak to the liver’s SUVmean. Cohen κ coefficient measured the rate of agreement of the DS assigned by the 2 radiologists. Receiver operating characteristic curves and the Youden index were used with the consensus DS to identify an optimal HLI cutoff value to distinguish PET-negative from PET-positive images. For each HLI cutoff value, sensitivity, specificity, and accuracy were calculated. Results: Interobserver agreement in DS was moderate (PET1: κ = 0.480, P < 0.001; PET2: κ = 0.428, P < 0.001). On PET1, the optimal cutoff values for HLImax were 1.19 for DS2/DS3 and 1.78 for DS3/DS4, and for HLIpeak were 1.19 for DS2/DS3 and 1.44 for DS3/DS4. On PET2 images, the corresponding values were 1.18, 1.58, 0.94, and 1.26. The most accurate predictor was HLImax on PET2 when scans were dichotomized as DS1–DS3 versus DS4–DS5 (accuracy, 92.8%; sensitivity, 100%; specificity, 91.8%). Conclusion: The HLI provides an objective, quantitative measure of disease response and thus may improve interreader agreement for lymphoma response assessments.
- 30 December 2024
- Published by Wolters Kluwer Health in Clinical Nuclear Medicine Open
- Vol. 1 (2)
Open AccessInterobserver Reliability of the 5-Point Deauville Score and SUV-Based Quantification of FDG PET/CT Scans in High-Risk Pediatric Hodgkin Lymphoma
Purpose: In lymphomas, the Deauville score (DS) standardizes 18F-FDG PET treatment response assessments. We assessed interobserver agreement of the DS and developed a quantitative PET-based parameter to minimize subjectivity of response assessments. Methods: PET scans performed within the Children’s Oncology Group AHOD0831 study were analyzed. One hundred one scans obtained after the first cycle of chemotherapy (PET1) and 83 obtained after the second cycle (PET2) were eligible for inclusion. Each scan was assigned a DS by 2 nuclear medicine radiologists and a Hodgkin lymphoma index (HLI), defined as the ratio of the tumor’s SUVmax or SUVpeak to the liver’s SUVmean. Cohen κ coefficient measured the rate of agreement of the DS assigned by the 2 radiologists. Receiver operating characteristic curves and the Youden index were used with the consensus DS to identify an optimal HLI cutoff value to distinguish PET-negative from PET-positive images. For each HLI cutoff value, sensitivity, specificity, and accuracy were calculated. Results: Interobserver agreement in DS was moderate (PET1: κ = 0.480, P < 0.001; PET2: κ = 0.428, P < 0.001). On PET1, the optimal cutoff values for HLImax were 1.19 for DS2/DS3 and 1.78 for DS3/DS4, and for HLIpeak were 1.19 for DS2/DS3 and 1.44 for DS3/DS4. On PET2 images, the corresponding values were 1.18, 1.58, 0.94, and 1.26. The most accurate predictor was HLImax on PET2 when scans were dichotomized as DS1–DS3 versus DS4–DS5 (accuracy, 92.8%; sensitivity, 100%; specificity, 91.8%). Conclusion: The HLI provides an objective, quantitative measure of disease response and thus may improve interreader agreement for lymphoma response assessments.
- Top Cited Papers
- 8 February 2024
- Published by Springer Nature in European Journal of Nuclear Medicine and Molecular Imaging
- Vol. 51 (7) , 1937-1954
TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images — a multi-center generalizability analysis
Purpose Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. Methods Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. Results Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. Conclusion TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.
- 8 February 2024
- Published by Springer Nature in European Journal of Nuclear Medicine and Molecular Imaging
- Vol. 51 (7) , 1937-1954
Top Cited PapersTMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images — a multi-center generalizability analysis
Purpose Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. Methods Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. Results Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. Conclusion TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.
- 1 June 2023
- Published by Springer Nature in Abdominal Radiology
- Vol. 48 (12) , 3643-3652
Distribution of prostate specific membrane antigen (PSMA) on PET-MRI in patients with and without ovarian cancer
Objectives Ovarian cancer is the most lethal cancer and future research needs to focus on the early detection and exploration of new therapeutic agents. The objectives of this proof-of-concept study are to assess the feasibility of PSMA 18F-DCFPyl PET/MR imaging for detecting ovarian cancer and to evaluate the PSMA distribution in patients with and without ovarian cancer. Methods This prospective pilot proof-of-concept study in patients with and without ovarian cancers occurred between October 2017 and January 2020. Patients were recruited from gynecologic oncology or hereditary ovarian cancer clinics, and underwent surgical removal of the uterus and ovaries for gynecologic indications. PSMA 18F-DCFPyl PET/MRI was obtained prior to standard of care surgery. Results Fourteen patients were scanned: four patients with normal ovaries, six patients with benign ovarian lesions, and four patients with malignant ovarian lesions. Tracer uptake in normal ovaries (SUVmax = 2.8 ± 0.4) was greater than blood pool (SUVmax = 1.8 ± 0.5, p < 0.0001). Tracer uptake in benign ovarian lesions (2.2 ± 1.0) did not differ significantly from blood pool (p = 0.331). Tracer uptake in ovarian cancer (SUVmax = 7.8 ± 3.8) was greater than blood pool (p < 0.0001), normal ovaries (p = 0.0014), and benign ovarian lesions (p = 0.005). Conclusion PET/MR imaging detected PSMA uptake in ovarian cancer, with little to no uptake in benign ovarian findings. These results are encouraging and further studies in a larger patient cohort would be useful to help determine the extent and heterogeneity of PSMA uptake in ovarian cancer patients.
- 1 June 2023
- Published by Springer Nature in Abdominal Radiology
- Vol. 48 (12) , 3643-3652
Distribution of prostate specific membrane antigen (PSMA) on PET-MRI in patients with and without ovarian cancer
Objectives Ovarian cancer is the most lethal cancer and future research needs to focus on the early detection and exploration of new therapeutic agents. The objectives of this proof-of-concept study are to assess the feasibility of PSMA 18F-DCFPyl PET/MR imaging for detecting ovarian cancer and to evaluate the PSMA distribution in patients with and without ovarian cancer. Methods This prospective pilot proof-of-concept study in patients with and without ovarian cancers occurred between October 2017 and January 2020. Patients were recruited from gynecologic oncology or hereditary ovarian cancer clinics, and underwent surgical removal of the uterus and ovaries for gynecologic indications. PSMA 18F-DCFPyl PET/MRI was obtained prior to standard of care surgery. Results Fourteen patients were scanned: four patients with normal ovaries, six patients with benign ovarian lesions, and four patients with malignant ovarian lesions. Tracer uptake in normal ovaries (SUVmax = 2.8 ± 0.4) was greater than blood pool (SUVmax = 1.8 ± 0.5, p < 0.0001). Tracer uptake in benign ovarian lesions (2.2 ± 1.0) did not differ significantly from blood pool (p = 0.331). Tracer uptake in ovarian cancer (SUVmax = 7.8 ± 3.8) was greater than blood pool (p < 0.0001), normal ovaries (p = 0.0014), and benign ovarian lesions (p = 0.005). Conclusion PET/MR imaging detected PSMA uptake in ovarian cancer, with little to no uptake in benign ovarian findings. These results are encouraging and further studies in a larger patient cohort would be useful to help determine the extent and heterogeneity of PSMA uptake in ovarian cancer patients.
- Open Access
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Supplementary Appendix from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
CONDOR study group, Supplementary tables, Study Protocol, & Medical Management Questionnaires
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Appendix from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
CONDOR study group, Supplementary tables, Study Protocol, & Medical Management Questionnaires
- Open Access
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Figure S1 from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
Negative or Equivocal Baseline Imaging
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Open AccessFigure S1 from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
Negative or Equivocal Baseline Imaging
- Open Access
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Data from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
Purpose:Current FDA-approved imaging modalities are inadequate for localizing prostate cancer biochemical recurrence (BCR). 18F-DCFPyL is a highly selective, small-molecule prostate-specific membrane antigen–targeted PET radiotracer. CONDOR was a prospective study designed to determine the performance of 18F-DCFPyL-PET/CT in patients with BCR and uninformative standard imaging.Experimental Design:Men with rising PSA ≥0.2 ng/mL after prostatectomy or ≥2 ng/mL above nadir after radiotherapy were eligible. The primary endpoint was correct localization rate (CLR), defined as positive predictive value with an additional requirement of anatomic lesion colocalization between 18F-DCFPyL-PET/CT and a composite standard of truth (SOT). The SOT consisted of, in descending priority (i) histopathology, (ii) subsequent correlative imaging findings, or (iii) post-radiation PSA response. The trial was considered a success if the lower bound of the 95% confidence interval (CI) for CLR exceeded 20% for two of three 18F-DCFPyL-PET/CT readers. Secondary endpoints included change in intended management and safety.Results:A total of 208 men with a median baseline PSA of 0.8 ng/mL (range: 0.2–98.4 ng/mL) underwent 18F-DCFPyL-PET/CT. The CLR was 84.8%–87.0% (lower bound of 95% CI: 77.8–80.4). A total of 63.9% of evaluable patients had a change in intended management after 18F-DCFPyL-PET/CT. The disease detection rate was 59% to 66% (at least one lesion detected per patient by 18F-DCFPyL-PET/CT by central readers).Conclusions:Performance of 18F-DCFPyL-PET/CT achieved the study’s primary endpoint, demonstrating disease localization in the setting of negative standard imaging and providing clinically meaningful and actionable information. These data further support the utility of 18F-DCFPyL-PET/CT to localize disease in men with recurrent prostate cancer.See related commentary by True and Chen, p. 3512
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Open AccessData from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
Purpose:Current FDA-approved imaging modalities are inadequate for localizing prostate cancer biochemical recurrence (BCR). 18F-DCFPyL is a highly selective, small-molecule prostate-specific membrane antigen–targeted PET radiotracer. CONDOR was a prospective study designed to determine the performance of 18F-DCFPyL-PET/CT in patients with BCR and uninformative standard imaging.Experimental Design:Men with rising PSA ≥0.2 ng/mL after prostatectomy or ≥2 ng/mL above nadir after radiotherapy were eligible. The primary endpoint was correct localization rate (CLR), defined as positive predictive value with an additional requirement of anatomic lesion colocalization between 18F-DCFPyL-PET/CT and a composite standard of truth (SOT). The SOT consisted of, in descending priority (i) histopathology, (ii) subsequent correlative imaging findings, or (iii) post-radiation PSA response. The trial was considered a success if the lower bound of the 95% confidence interval (CI) for CLR exceeded 20% for two of three 18F-DCFPyL-PET/CT readers. Secondary endpoints included change in intended management and safety.Results:A total of 208 men with a median baseline PSA of 0.8 ng/mL (range: 0.2–98.4 ng/mL) underwent 18F-DCFPyL-PET/CT. The CLR was 84.8%–87.0% (lower bound of 95% CI: 77.8–80.4). A total of 63.9% of evaluable patients had a change in intended management after 18F-DCFPyL-PET/CT. The disease detection rate was 59% to 66% (at least one lesion detected per patient by 18F-DCFPyL-PET/CT by central readers).Conclusions:Performance of 18F-DCFPyL-PET/CT achieved the study’s primary endpoint, demonstrating disease localization in the setting of negative standard imaging and providing clinically meaningful and actionable information. These data further support the utility of 18F-DCFPyL-PET/CT to localize disease in men with recurrent prostate cancer.See related commentary by True and Chen, p. 3512
- Open Access
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Figure S1 from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
Negative or Equivocal Baseline Imaging
- 31 March 2023
- Published by American Association for Cancer Research (AACR)
Open AccessFigure S1 from Diagnostic Performance of 18F-DCFPyL-PET/CT in Men with Biochemically Recurrent Prostate Cancer: Results from the CONDOR Phase III, Multicenter Study
Negative or Equivocal Baseline Imaging