Perspectives on technology: Prostate Imaging-Reporting and Data System (PI-RADS) interobserver variability.

OBJECTIVES: To explore the topic of Prostate Imaging-Reporting and Data System (PI-RADS) interobserver variability, including a discussion of major sources, mitigation approaches, and future directions. METHODS: A narrative review of PI-RADS interobserver variability. RESULTS: PI-RADS was developed in 2012 to set technical standards for prostate magnetic resonance imaging (MRI), reduce interobserver variability at interpretation, and improve diagnostic accuracy in the MRI-directed diagnostic pathway for detection of clinically significant prostate cancer. While PI-RADS has been validated in selected research cohorts with prostate cancer imaging experts, subsequent prospective studies in routine clinical practice demonstrate wide variability in diagnostic performance. Radiologist and biopsy operator experience are the most important contributing drivers of high-quality care among multiple interrelated factors including variability in MRI hardware and technique, image quality, and population and patient-specific factors such as prostate cancer disease prevalence. Iterative improvements in PI-RADS have helped flatten the curve for novice readers and reduce variability. Innovations in image quality reporting, administrative and organisational workflows, and artificial intelligence hold promise in improving variability even further. CONCLUSION: Continued research into PI-RADS is needed to facilitate benchmark creation, reader certification, and independent accreditation, which are systems-level interventions needed to uphold and maintain high-quality prostate MRI across entire populations.

Patterns of Metastatic Disease in Patients with Cancer Derived from Natural Language Processing of Structured CT Radiology Reports over a 10-year Period.

Background Patterns of metastasis in cancer are increasingly relevant to prognostication and treatment planning but have historically been documented by means of autopsy series. Purpose To show the feasibility of using natural language processing (NLP) to gather accurate data from radiology reports for assessing spatial and temporal patterns of metastatic spread in a large patient cohort. Materials and Methods In this retrospective longitudinal study, consecutive patients who underwent CT from July 2009 to April 2019 and whose CT reports followed a departmental structured template were included. Three radiologists manually curated a sample of 2219 reports for the presence or absence of metastases across 13 organs; these manually curated reports were used to develop three NLP models with an 80%-20% split for training and test sets. A separate random sample of 448 manually curated reports was used for validation. Model performance was measured by accuracy, precision, and recall for each organ. The best-performing NLP model was used to generate a final database of metastatic disease across all patients. For each cancer type, statistical descriptive reports were provided by analyzing the frequencies of metastatic disease at the report and patient levels. Results In 91 665 patients (mean age ± standard deviation, 61 years ± 15; 46 939 women), 387 359 reports were labeled. The best-performing NLP model achieved accuracies from 90% to 99% across all organs. Metastases were most frequently reported in abdominopelvic (23.6% of all reports) and thoracic (17.6%) nodes, followed by lungs (14.7%), liver (13.7%), and bones (9.9%). Metastatic disease tropism is distinct among common cancers, with the most common first site being bones in prostate and breast cancers and liver among pancreatic and colorectal cancers. Conclusion Natural language processing may be applied to cancer patients' CT reports to generate a large database of metastatic phenotypes. Such a database could be combined with genomic studies and used to explore prognostic imaging phenotypes with relevance to treatment planning. © RSNA, 2021 Online supplemental material is available for this article.

Abdominal imaging findings on computed tomography in patients acutely infected with SARS-CoV-2: what are the findings?

OBJECTIVES: To investigate what findings are new on contrast-enhanced abdominopelvic CT in patients infected with SARS-CoV-2. METHODS: Contrast-enhanced CT of the abdomen and pelvis of patients with COVID-19 at a tertiary oncologic center acquired over a 2-month period were reviewed independently by two readers and scored for new imaging abnormalities compared with a prior scan. CT scans were included if the study was performed between - 3 and 45 days from the time of COVID-19 diagnosis. Clinical information was gathered from the medical records. RESULTS: A total of 63 patients (34 male, 29 female; mean age 60.6 years, range 24.4-85.0 years) were included in this retrospective cross-sectional study. Aside from new ground glass opacities seen at the lung bases (29/63, 46.0%), the most common findings were new thickening of the stomach, small bowel or colon or fluid-filled colon (14/63, 22.2%), new small volume ascites (7/63, 14.3%), gallbladder distention in those without prior cholecystectomy (3/43, 7.0%), and single cases each of acute pancreatitis (1/63, 1.6%) as well as new portal vein thrombosis (1/63, 1.6%). CONCLUSION: Aside from lung base ground glass opacities, the most common new imaging abnormality on abdominopelvic CT in patients with COVID-19 finding in our cohort was abnormalities of the gastrointestinal tract, followed by small volume ascites, gallbladder distention, and isolated cases of pancreatitis and portal vein thrombosis. These findings overlap with those previously reported that did not have a prior scan for comparison, and provide supportive evidence that some of these findings may be related to SARS-CoV-2 infection.

The relationship between adrenal incidentalomas and mortality risk.

OBJECTIVE: To determine all-cause mortality risk in patients with and without adrenal incidentaloma. METHODS: Retrospective cohort study of patients with CT abdomen performed within 24 h of emergency room presentation at an academic medical center from January 1, 2005, to December 31, 2009, without history of adrenal disease, adrenal lab testing, or cancer. Incidentaloma cohort identified by database query of imaging reports followed by manual review and matched to no-nodule controls at 3:1 on age ± 1 year and exam date ± 3 months. Mortality ascertained by in-hospital deaths and National Death Index query. Survival analysis performed with Kaplan-Meier curves and Cox proportional hazards models. RESULTS: Among 42,575 adults with abdominal CT exams, 969 adrenal incidentaloma patients and 2907 no-nodule controls were identified. All 3876 individuals entered survival analysis with 31,182 person-years at risk (median follow-up 8.9 years [IQR, 6.9-10.7]). All-cause mortality was significantly higher among those with adrenal incidentalomas (353/969, 36.4%) compared with those without (919/2907, 31.6%; mortality difference 7.6 per 1000 person-years; multivariable-adjusted hazard ratio [aHR] 1.14; 95% CI, 1.003-1.29). Exploratory analyses, limited by missing covariates, found that adrenal incidentalomas were associated with significantly increased incidence of malignancy (aHR 1.61; 95% CI, 1.22-2.12), diabetes (aHR 1.43; 95% CI, 1.18-1.71), heart failure (aHR 1.32; 95% CI, 1.07-1.63), peripheral vascular disease (aHR 1.28; 95% CI, 1.95-1.56), renal disease (aHR 1.21; 95% CI, 1.01-1.44), and chronic pulmonary disease (aHR 1.22; 95% CI, 1.01-1.46) compared with controls. CONCLUSIONS: Adrenal incidentalomas are associated with increased mortality and may represent a clinically valuable biomarker. KEY POINTS: • Adrenal incidentalomas are associated with increased mortality. • Adrenal incidentaloma size is not predictive of mortality. • On exploratory analyses, adrenal incidentalomas are associated with chronic illnesses.

Evaluating the Impact of Annual Resident Retreats on Radiology Resident Wellbeing.

RATIONALE AND OBJECTIVES: To investigate the effect of resident-run wellness retreats on measures of stress, resilience, and cohesion amongst radiology residents. MATERIALS AND METHODS: All diagnostic and interventional radiology residents in a single academic medical center were invited to participate in a wellness-focused off-campus retreat. Three retreats were held on an annual basis from 2018 to 2022 and included physical exercises, nutritional wellness, and team-building activities. Participants were surveyed pre- and post- retreat attendance focusing on the following domains: Perceived Stress Scale, higher scores indicating higher levels of perceived stress; Brief Resilience Scale, higher scores indicating greater resilience; Cohesion Scale, higher scores indicating greater cohesion. Paired t-tests and linear regression models were used to compare mean Perceived Stress, Brief Resilience, and Cohesion Scales pre- and post-retreat across covariates. RESULTS: Of 78 total retreat attendances, 75 residents (96%) completed pre- and/or post-surveys. Study population was 51% female and 49% male, with mean age 29 ± 2.8 years. Comparing post- with pre-surveys, retreat attendance was associated with a significant reduction in mean Perceived Stress (12.7 vs 16.3; 95% CI 2.1-5.3) and significant increase in Brief Resilience (3.9 vs 3.7; 95% CI 0.05-0.34) and Cohesion (33 vs 27; 95% CI 4.3-6.7). First year residents experienced the greatest increase in Cohesion compared to fourth year residents (p < 0.001). Pre-retreat cohesion was significantly lower in 2022 than in 2018-2019 (26.6 vs 28.7 vs 28.6; p = 0.04), with stress and resilience not significantly different by year. CONCLUSION: An annual resident retreat decreased stress, while increasing resilience and cohesion within a radiology residency, supporting retreats as a potentially viable intervention to promote physician well-being.

Pancreatic Cystic Lesions: Imaging Techniques and Diagnostic Features.

The detection of incidental pancreatic cystic lesions has increased over time. It is crucial to separate benign from potentially malignant or malignant lesions to guide management and reduce morbidity and mortality. The key imaging features used to fully characterize cystic lesions are optimally assessed by contrast-enhanced magnetic resonance imaging/magnetic resonance cholangiopancreatography, with pancreas protocol computed tomography offering a complementary role. While some imaging features have high specificity for a particular diagnosis, overlapping imaging features between diagnoses may require further investigation with follow-up diagnostic imaging or tissue sampling.

A Novel Hands-on Approach Towards Teaching Diagnostic Radiology Residents MRI Scanning and Physics.

RATIONALE AND OBJECTIVES: Traditional approaches towards teaching magnetic resonance imaging (MRI) scanning and physics have limitations that a hands-on course may help overcome. A dedicated week of MRI instruction may help improve radiology resident confidence and competence. Additional benefits may include improved physician-technologist communication and accelerated mastery of MRI safety. MATERIALS AND METHODS: Surveys and tests were approved by our Program Evaluation Committee and administered at the beginning and at the end of this one-week course. The course consisted of protected reading time as well as practice scanning with a research magnet and assisting with clinical scanning under the close supervision of a licensed MRI technologist. Eighteen senior residents (nine third-year and nine fourth-year) participated in this course during its first year. RESULTS: Few residents had previous experience with MRI physics, scanning, or research prior to residency. After this course, mean resident confidence increased by 0.47 points (3.33 vs 2.86; p=0.01) on a five-point Likert scale. Understanding of MRI physics, as measured by pre- and post-tests, increased by 22% (0.72 vs 0.50; p<0.01), corresponding to a large effect size of 1.29 (p<0.001). Resident feedback reported that this course was efficacious (5/5), engaging (4.9/5), and had optimal faculty oversight. The most highly rated component of the course was the opportunity to experiment with the research MR scanner (5/5). CONCLUSION: A dedicated week of MRI education was highly rated by residents and associated with improvements in confidence and understanding, suggesting a positive correlation between confidence and competence. Additional metrics, such as trends in scores on the American Board of Radiology's Core Examination over the next several years, may further support the apparent benefits of this hands-on MR course.

Natural Language Processing of Computed Tomography Reports to Label Metastatic Phenotypes With Prognostic Significance in Patients With Colorectal Cancer.

PURPOSE: Natural language processing (NLP) applied to radiology reports can help identify clinically relevant M1 subcategories of patients with colorectal cancer (CRC). The primary purpose was to compare the overall survival (OS) of CRC according to American Joint Committee on Cancer TNM staging and explore an alternative classification. The secondary objective was to estimate the frequency of metastasis for each organ. METHODS: Retrospective study of CRC who underwent computed tomography (CT) chest, abdomen, and pelvis between July 1, 2009, and March 26, 2019, at a tertiary cancer center, previously labeled for the presence or absence of metastasis by an NLP prediction model. Patients were classified in M0, M1a, M1b, and M1c (American Joint Committee on Cancer), or an alternative classification on the basis of the metastasis organ number: M1, single; M2, two; M3, three or more organs. Cox regression models were used to estimate hazard ratios; Kaplan-Meier curves were used to visualize survival curves using the two M1 subclassifications. RESULTS: Nine thousand nine hundred twenty-eight patients with a total of 48,408 CT chest, abdomen, and pelvis reports were included. On the basis of NLP prediction, the median OS of M1a, M1b, and M1c was 4.47, 1.72, and 1.52 years, respectively. The median OS of M1, M2, and M3 was 4.24, 2.05, and 1.04 years, respectively. Metastases occurred most often in liver (35.8%), abdominopelvic lymph nodes (32.9%), lungs (29.3%), peritoneum (22.0%), thoracic nodes (19.9%), bones (9.2%), and pelvic organs (7.5%). Spleen and adrenal metastases occurred in < 5%. CONCLUSION: NLP applied to a large radiology report database can identify clinically relevant metastatic phenotypes and be used to investigate new M1 substaging for CRC. Patients with three or more metastatic disease organs have the worst prognosis, with an OS of 1 year.

Comparison of extracolonic findings and clinical outcomes in a screening and diagnostic CT colonography population.

PURPOSE: To compare the distribution of extracolonic findings and clinical outcomes between screening and diagnostic CT colonography (CTC) populations. METHODS: 388 consecutive patients (369 men, 19 women; mean ± SD age 67.8 ± 10 years) who underwent first-time CTC (4/2011-4/2017) at a Veteran's Affairs Medical Center were divided into screening (asymptomatic) or diagnostic (symptomatic) cohorts based on CTC indication. CTC reporting and data system E-scores for extracolonic findings were retrospectively assigned based on prospective CTC radiologic reports. Multinomial logistic regression was used to examine the association between E-scores and CTC indication. Electronic medical records of all patients with E3 or E4 scores were reviewed (median follow-up 2.8 years) to determine clinical outcomes. RESULTS: 68% (262/388) underwent screening and 32% (126/388) diagnostic CTC. 7.2% (28/388) had extracolonic findings considered potentially significant (E4), 4.4% (17/388) had indeterminate but likely unimportant findings (E3), and 88.4% (347/388) had normal or unimportant findings (E1 or E2). E-scores were not significantly different between screening and diagnostic CTC when adjusted for age, gender, and prior imaging (p = 0.44). 4.6% (12/262) of patients with E3/E4 findings in the screening cohort demonstrated clinically significant outcomes, compared with 4.0% (5/126) in the diagnostic cohort, including a total of three extracolonic malignancies (0.8%) and three abdominal aortic aneurysms (0.8%). 4.6% (18/388) underwent follow-up imaging studies to confirm a benign outcome after detection of a category E3/E4 finding. CONCLUSIONS: The distribution of extracolonic findings and clinical outcomes were not statistically significantly different between screening and diagnostic CTC populations.

pH-responsive artemisinin derivatives and lipid nanoparticle formulations inhibit growth of breast cancer cells in vitro and induce down-regulation of HER family members.

Artemisinin (ART) dimers show potent anti-proliferative activities against breast cancer cells. To facilitate their clinical development, novel pH-responsive artemisinin dimers were synthesized for liposomal nanoparticle formulations. A new ART dimer was designed to become increasingly water-soluble as pH declines. The new artemisinin dimer piperazine derivatives (ADPs) remained tightly associated with liposomal nanoparticles (NPs) at neutral pH but were efficiently released at acidic pH's that are known to exist within solid tumors and organelles such as endosomes and lysosomes. ADPs incorporated into nanoparticles down regulated the anti-apoptotic protein, survivin, and cyclin D1 when incubated at low concentrations with breast cancer cell lines. We demonstrate for the first time, for any ART derivative, that ADP NPs can down regulate the oncogenic protein HER2, and its counterpart, HER3 in a HER2+ cell line. We also show that the wild type epidermal growth factor receptor (EGFR or HER1) declines in a triple negative breast cancer (TNBC) cell line in response to ADP NPs. The declines in these proteins are achieved at concentrations of NP109 at or below 1 µM. Furthermore, the new artemisinin derivatives showed improved cell-proliferation inhibition effects compared to known dimer derivatives.