Combined artificial intelligence and radiologist model for predicting rectal cancer treatment response from magnetic resonance imaging: an external validation study.

PURPOSE: To evaluate an MRI-based radiomic texture classifier alone and combined with radiologist qualitative assessment in predicting pathological complete response (pCR) using restaging MRI with internal training and external validation. METHODS: Consecutive patients with locally advanced rectal cancer (LARC) who underwent neoadjuvant therapy followed by total mesorectal excision from March 2012 to February 2016 (Memorial Sloan Kettering Cancer Center/internal dataset, n = 114, 41% female, median age = 55) and July 2014 to October 2015 (Instituto do Câncer do Estado de São Paulo/external dataset, n = 50, 52% female, median age = 64.5) were retrospectively included. Two radiologists (R1, senior; R2, junior) independently evaluated restaging MRI, classifying patients (radiological complete response vs radiological partial response). Model A (n = 33 texture features), model B (n = 91 features including texture, shape, and edge features), and two combination models (model A + B + R1, model A + B + R2) were constructed. Pathology served as the reference standard for neoadjuvant treatment response. Comparison of the classifiers' AUCs on the external set was done using DeLong's test. RESULTS: Models A and B had similar discriminative ability (P = 0.3; Model B AUC = 83%, 95% CI 70%-97%). Combined models increased inter-reader agreement compared with radiologist-only interpretation (κ = 0.82, 95% CI 0.70-0.89 vs k = 0.25, 95% CI 0.11-0.61). The combined model slightly increased junior radiologist specificity, positive predictive value, and negative predictive values (93% vs 90%, 57% vs 50%, and 91% vs 90%, respectively). CONCLUSION: We developed and externally validated a combined model using radiomics and radiologist qualitative assessment, which improved inter-reader agreement and slightly increased the diagnostic performance of the junior radiologist in predicting pCR after neoadjuvant treatment in patients with LARC.

Value of Primary Rectal Tumor PET/MRI in the Prediction of Synchronic Metastatic Disease.

PURPOSE: To analyze the associations between positron emission tomography (PET)/magnetic resonance imaging (MRI) features for primary rectal tumors and metastases. PROCEDURES: Between November 2016 and April 2018, 101 patients with rectal adenocarcinoma were included in this prospective study (NCT02537340) for whole-body PET/MRI for baseline staging. Two readers analyzed the PET/MRI; they assessed the semiquantitative PET features of the primary tumor and the N- and M-stages. Another reader analyzed the MRI features for locoregional staging. The reference standard for confirming metastatic disease was biopsy or imaging follow-up. Non-parametric tests were used to compare the PET/MRI features of the participants with or without metastatic disease. Binary logistic regression was used to evaluate the associations between the primary tumor PET/MRI features and metastatic disease. RESULTS: A total of 101 consecutive participants (median age 62 years; range: 33-87 years) were included. Metastases were detected in 35.6% (36 of 101) of the participants. Among the PET/MRI features, higher tumor lesion glycolysis (352.95 vs 242.70; P = .46) and metabolic tumor volume (36.15 vs 26.20; P = .03) were more frequent in patients with than in those without metastases. Additionally, patients with metastases had a higher incidence of PET-positive (64% vs 32%; P = .009) and MRI-positive (56% vs 32%; P = .03) mesorectal lymph nodes, extramural vascular invasion (86% vs 49%; P > .001), and involvement of mesorectal fascia (64% vs 42%; P = .04); there were also differences between the mrT stages of these two groups (P = .008). No differences in the maximum standardized uptake values for the primary tumors in patients with and without metastases were observed (18.9 vs 19.1; P = .56). Multivariable logistic regression showed that extramural vascular invasion on MRI was the only significant predictor (adjusted odds ratio, 3.8 [95% CI: 1.1, 13.9]; P = .001). CONCLUSION: PET/MRI facilitated the identification of participants with a high risk of metastatic disease, though these findings were based mainly on MRI features.

Diagnostic accuracy of FDG-PET/MRI versus pelvic MRI and thoracic and abdominal CT for detecting synchronous distant metastases in rectal cancer patients.

PURPOSE: We compared the diagnostic accuracy of detecting distant metastases for baseline rectal cancer staging between PET/MRI and conventional staging (CS). MATERIALS AND METHODS: This prospective study from November 2016 to April 2018 included 101 rectal adenocarcinoma patients for primary staging. These patients underwent whole-body PET/MRI in addition to CS (pelvic MRI and thoracic and abdominal contrast-enhanced CT). Different readers analyzed CS and PET/MRI findings for primary tumor, nodal, and metastatic staging. The presence, number, and location of metastases were recorded according to the organ involved (non-regional lymph nodes (LNs), liver, lungs, or others). Lesions were defined as positive, negative, or indeterminate. The number of lesions per organ was limited to 10. The McNemar test was used to compare the accuracies. RESULTS: PET/MRI exhibited a higher accuracy in detecting metastatic disease than CS in all patients (88.4% vs. 82.6%, p = 0.003) and in patients with extramural vascular invasion (EMVI) (88.9% vs. 85.5%, p = 0.013). The detection rate of PET/MRI was superior to that of CS for all lesions [84.1% vs. 68.9%, p = 0.001], as well as those in the liver (89.2% vs. 84.2%), non-regional LNs (90.0% vs. 36.7%), and lungs (76.4% vs. 66.9%). PET/MRI correctly classified 19/33 (57.5%) patients with indeterminate lesions on CS. CONCLUSION: PET/MRI yields higher accuracy than CS for detecting distant synchronous metastases in the baseline staging of patients with rectal cancer and EMVI. PET/MRI exhibited a higher detection rate than CS for identifying non-regional LNs, hepatic lesions, and pulmonary lesions as well as correctly classifying patients with indeterminate lesions. TRIAL REGISTRATION: NCT02537340.

Imaging Findings in Transgender Patients after Gender-affirming Surgery.

Gender-affirming surgeries expand the options for physical transition among transgender patients, those whose gender identity is incongruent with the sex assigned to them at birth. Growing medical insight, increasing public acceptance, and expanding insurance coverage have improved the access to and increased the demand for gender-affirming surgeries in the United States. Procedures for transgender women, those patients with feminine gender identity, include breast augmentation using implants and genital reconstruction with vaginoplasty. Some transgender women receive medically unapproved silicone injections for breast augmentation or other soft-tissue contouring procedures that can lead to disfigurement, silicone pulmonary embolism, systemic reactions, and even death. MRI is preferred over CT for postvaginoplasty evaluation given its superior tissue contrast resolution. Procedures for transgender men, patients with a masculine gender identity, include chest masculinization (mastectomy) and genital reconstruction (phalloplasty or metoidioplasty, scrotoplasty, and erectile device implantation). Urethrography is the standard imaging modality performed to evaluate neourethral patency and other complications, such as leaks and fistulas. Despite a sizeable growth in the surgical literature about gender-affirming surgeries and their outcomes, detailed descriptions of the imaging features following these surgeries remain sparse. Radiologists must be aware of the wide variety of anatomic and pathologic changes unique to patients who undergo gender-affirming surgeries to ensure accurate imaging interpretation. Online supplemental material is available for this article. ©RSNA, 2019.

Inconsistency prevents the valuable synergism of explanatory and pragmatic trails.

OBJECTIVE: To assess review articles on pragmatic trials in order to describe how authors define the aim of this type of study, how comprehensive methodological topics are covered, and which topics are most valued by authors. METHODS: Review articles were selected from Medline Database, based on the expression "pragmatic trial" in the titles. Five trained medical students evaluated the articles, based on a list of 15 self-explanatory methodological topics. Each article was evaluated regarding topics covered. Baseline statements on the aim of pragmatic trials were derived. RESULTS: Among 22 articles identified, there was general agreement that the aim of a pragmatic trial is to evaluate if the intervention works under real-world conditions. The mean number of methodological topics addressed by each article was 7.6 ± 3.1. Only one article covered all 15 topics, three articles (14%) responded to at least 75% of topics and 13 articles (59%) mentioned at least 50% of the topics. The relative frequency each of the 15 topics was cited by articles had a mean of 50% ± 25%. No topic was addressed by all articles, only three (20%) were addressed by more than 75% of articles. CONCLUSIONS: There is agreement on the different aims of explanatory and pragmatic trials. But there is a large variation on methodological topics used to define a pragmatic trial, which led to inconsistency in defining the typical methodology of a pragmatic trial.