
A whistleblower lawsuit has alleged that Cincinnati-based imaging services provider ProScan submitted thousands of reimbursement claims to Medicare, Medicaid, and Tricare for MRI interpretations that were performed by physician assistants instead of radiologists.
The lawsuit was unsealed by a federal judge in Cincinnati on September 12. It was filed in the U.S. District Court in Cincinnati in October 2017 by Jason Taylor, a former ProScan radiology assistant, and Dr. Peter Rothschild, a California radiologist who reviewed numerous ProScan reports after receiving calls from doctors who had "grave concerns" with the results, according to a September 13 report by WCPO-TV. The suit alleges that ProScan used physician assistants to "ghost read" MR images, which resulted in frequent misdiagnoses and diagnoses that were missed.
"If these millions of MRIs were reread by board-certified radiologists, the true scale of this tragedy would become clear," the lawsuit states.
The suit details allegations from Rothschild regarding MRI cases read at ProScan, including a missed diagnosis of severe spinal cord injuries and a complete tear of a ligament that stabilizes the neck in a jockey after a horse fell on him.
WCPO-TV reported that Taylor was hired to work in ProScan's marketing and sales department in 2015. The lawsuit alleges that ProScan CEO and Medical Director Dr. Stephen Pomeranz then tried to recruit him as a "ghost reader" for ProScan by having him attend a one-year, uncertified, secret training program.
The U.S. Department of Justice (DOJ) declined to intervene in the case, but it allowed Taylor and Rothschild to litigate on behalf of the U.S., WCPO-TV reported. The alleged whistleblowers are suing ProScan Imaging, which operates 25 freestanding imaging centers in seven states, and ProScan's teleradiology division ProScan Reading, which reads approximately 2,000 MRI scans per day for 500 hospitals and imaging centers nationwide, according to the lawsuit. Pomeranz and Dr. Malcolm Shupeck, associate director of ProScan and director of fellowship administration for the ProScan Imaging Education Foundation, are also named in the suit, according to WCPO-TV.
In addition, the lawsuit alleges that it's impossible for ProScan's 35 board-certified radiologists to read 350,000 studies per year, as it advertised in 2018, the TV station also reported.
ProScan insists, however, that all of its radiology reports are reviewed and finalized by licensed, board-certified physicians. The WCPO-TV article included a statement from Pomeranz and ProScan President Michael O'Brien:
ProScan was contacted by the DOJ over a year ago with a request for information related to the government's investigation of the allegations of the lawsuit. ProScan responded to the DOJ's requests, shared all requested information, and fully cooperated with the investigation. After the DOJ's thorough review and consideration of all requested information, the DOJ declined to pursue the case further.
Pomeranz and O'Brien added that the company has always conducted "business within strict adherence to compliant policies and procedures and will continue to do so." In addition, they stated that the allegations "are devoid of merit" and that the company is confident the lawsuit will be dismissed.



![Overview of the study design. (A) The fully automated deep learning framework was developed to estimate body composition (BC) (defined as subcutaneous adipose tissue [SAT] in liters; visceral adipose tissue [VAT] in liters; skeletal muscle [SM] in liters; SM fat fraction [SMFF] as a percentage; and intramuscular adipose tissue [IMAT] in deciliters) from MRI. The fully automated framework comprised one model (model 1) to quantify different BC measures (SAT, VAT, SM, SMFF, and IMAT) as three-dimensional (3D) measures from whole-body MRI scans. The second model (model 2) was trained to identify standardized anatomic landmarks along the craniocaudal body axis (z coordinate field), which allowed for subdividing the whole-body measures into different subregions typically examined on clinical routine MRI scans (chest, abdomen, and pelvis). (B) BC was quantified from whole-body MRI in over 66,000 individuals from two large population-based cohort studies, the UK Biobank (UKB) (36,317 individuals) and the German National Cohort (NAKO) (30,291 individuals). Bar graphs show age distribution by sex and cohort. BMI = body mass index. (C) After the performance assessment of the fully automated framework, the change in BC measures, distributions, and profiles across age decades were investigated. Age-, sex-, and height-adjusted body composition reference curves were calculated and made publicly available in a web-based z-score calculator (https://circ-ml.github.io).](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/05/body-comp.XgAjTfPj1W.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)






![Overview of the study design. (A) The fully automated deep learning framework was developed to estimate body composition (BC) (defined as subcutaneous adipose tissue [SAT] in liters; visceral adipose tissue [VAT] in liters; skeletal muscle [SM] in liters; SM fat fraction [SMFF] as a percentage; and intramuscular adipose tissue [IMAT] in deciliters) from MRI. The fully automated framework comprised one model (model 1) to quantify different BC measures (SAT, VAT, SM, SMFF, and IMAT) as three-dimensional (3D) measures from whole-body MRI scans. The second model (model 2) was trained to identify standardized anatomic landmarks along the craniocaudal body axis (z coordinate field), which allowed for subdividing the whole-body measures into different subregions typically examined on clinical routine MRI scans (chest, abdomen, and pelvis). (B) BC was quantified from whole-body MRI in over 66,000 individuals from two large population-based cohort studies, the UK Biobank (UKB) (36,317 individuals) and the German National Cohort (NAKO) (30,291 individuals). Bar graphs show age distribution by sex and cohort. BMI = body mass index. (C) After the performance assessment of the fully automated framework, the change in BC measures, distributions, and profiles across age decades were investigated. Age-, sex-, and height-adjusted body composition reference curves were calculated and made publicly available in a web-based z-score calculator (https://circ-ml.github.io).](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/05/body-comp.XgAjTfPj1W.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)








