Applying Natural Language Processing to Single-Report Prediction of Metastatic Disease Response Using the OR-RADS Lexicon.

Generating Real World Evidence (RWE) on disease responses from radiological reports is important for understanding cancer treatment effectiveness and developing personalized treatment. A lack of standardization in reporting among radiologists impacts the feasibility of large-scale interpretation of disease response. This study examines the utility of applying natural language processing (NLP) to the large-scale interpretation of disease responses using a standardized oncologic response lexicon (OR-RADS) to facilitate RWE collection. Radiologists annotated 3503 retrospectively collected clinical impressions from radiological reports across several cancer types with one of seven OR-RADS categories. A Bidirectional Encoder Representations from Transformers (BERT) model was trained on this dataset with an 80-20% train/test split to perform multiclass and single-class classification tasks using the OR-RADS. Radiologists also performed the classification to compare human and model performance. The model achieved accuracies from 95 to 99% across all classification tasks, performing better in single-class tasks compared to the multiclass task and producing minimal misclassifications, which pertained mostly to overpredicting the equivocal and mixed OR-RADS labels. Human accuracy ranged from 74 to 93% across all classification tasks, performing better on single-class tasks. This study demonstrates the feasibility of the BERT NLP model in predicting disease response in cancer patients, exceeding human performance, and encourages the use of the standardized OR-RADS lexicon to improve large-scale prediction accuracy.

Visceral adiposity in cirrhosis: Association with disease severity and impact of liver transplantation.

BACKGROUND: Changes in adipose tissue distribution in liver cirrhosis are poorly characterized and may affect clinical outcomes. METHODS: Adult liver transplant (LT) January 2008-August 2017 recipients with abdominal MRI within 6 months pre-LT were retrospectively assessed. Visceral adipose tissue, subcutaneous adipose tissue, and skeletal muscle area (cm2) were determined at L3. Visceral-to-subcutaneous adipose tissue ratio (VSR) was used to define relative adipose distribution, stratified by sex. Correlation was tested with Pearson. Body composition measures were compared by Child-Turcotte-Pugh (CTP) class, before and after LT, and evaluated as predictors of clinical outcomes. RESULTS: A total of 318 patients were studied. Mean age was 56 years, 33.64% were female, and 47.80% had CTP C cirrhosis. CTP C was associated with a 0.42-point increase in VSR compared with CTP A (95% CI = 0.13-0.71, p < 0.01), adjusting for age, sex, diabetes, and HCC. Among the 79 (24.84%) patients with repeat MRI 1-2 years after LT, VSR significantly improved from before LT (1.31 vs. 0.95, p < 0.01). In adjusted analysis, CTP C was associated with a 0.86-point decrease in post-LT VSR compared with pre-LT VSR (95% CI = -1.27 to -0.44, p < 0.01). Body mass index poorly correlated with VSR before and after LT. Elevated pre-LT VSR trended toward an association with a 7.17-point decrease in pre-LT glomerular filtration rate (95% CI = -14.35 to -0.02, p = 0.05), adjusting for CTP C, age, sex, diabetes, hypertension, pre-LT sarcopenia, and hepatocellular carcinoma. Elevated pre-LT VSR did not affect 3-year post-LT mortality (log-rank p = 0.24). CONCLUSIONS: Poorly represented by body mass index, visceral adiposity is increased in cirrhosis and is associated with CTP class. However, this adipose redistribution may be modifiable by LT.

One-step synthesis of magnetic fly ash composites for methylene blue removal: batch and column study.

This study focuses on the synthesis of magnetic fly ash composites and its application for the removal of methylene blue (MB) from wastewater. By-product of oil power plants, oil fly ash, was treated with magnetic nanoparticles after chemical surface modification and dubbed modified fly ash (MFA). MFA was characterized by X-ray fluorescence, diffractogram analysis, scanning electron microscopy, energy-dispersive X-ray (EDX), and Fourier transform infrared (FT-IR) and N2 physisorption. MB (methylene blue) was removed from an aqueous solution using the response surface modelling (RSM) technique, which was used for optimization reasons. All four independent factors were investigated to see how they affected the removal process: adsorbent dosage; contact time; pH; and beginning dye concentration. The rate of MB removal was strongly influenced by the pH of the solution. The Langmuir and Freundlich models were used to examine equilibrium data A for MB adsorption onto the MFA in linear and nonlinear forms. Langmuir gave a better fit. The adsorption kinetics shown by increased kinetic statistics were better characterized by a pseudo-second-order MFA model. As far as thermodynamic characteristics go, adsorption is endothermic and occurs spontaneously. It has been proven that MFA may be used as an adsorbent to remove MB dye with high efficiency, and the quadratic model has been proved to be statistically significant.

Association of Patient and Imaging-Related Factors With False Negative MRI-Targeted Prostate Biopsies of Suspicious PI-RADS 4 and 5 Lesions.

OBJECTIVE: To investigate specific imaging and patient-related factors associated with a false negative (FN) MRI-targeted prostate fusion biopsies (TBx) of suspicious MRI lesions. METHODS: Retrospective study of men with PI-RADS 4 or 5 lesions November, 2015-December 2020 with TBx and systematic biopsy (SBx) performed. Only FN and true positive (TP) targeted lesions were included. FN biopsy was defined as a negative TBx with a positive systematic core in the ROI or perilesional sextant. Logistic regression was used to determine the association of patient and imaging-specific factors with the probability of a FN TBx. RESULTS: 361 PI-RADS 4 or 5 lesions in 304 patients, including 67 FN (19%) and 294 TP (81%) were included. There was a significant inverse association between lesion size (OR: 0.94, P-value: .02), presence of a suspicious DRE (OR: 0.36, P-value: .02) and PSA density (OR: 0.01, P-value: .004) on the probability of obtaining a FN TBx. There was no association between age, biopsy indication, use of an enema before MRI, prostate size, or discrepant US and MRI segmentation volumes on the probability of a FN TBx. CONCLUSION: In this cohort, SBx detected 19% of csPCa missed on TBx. Smaller PI-RADS 4/5 lesions, lower PSAD values, and a normal DRE were all associated with an increased probability of a FN TBx.

Molecular identification of two entomopathogenic fungus Clonostachys rosea strains and their efficacy against two aphid species in Iraq.

BACKGROUND: The green peach aphid, Myzus persicae Sulzer, and the bean aphid, Aphis fabae Scopoli (both Hemiptera: Aphididae), are serious pests of greenhouse vegetable crops in Iraq and other regions of the globe. In this study, two morphological identical isolates (AA80 and AA82) of the entomopathogenic fungus Clonostachys rosea Schroers (Hypocreales: Bionectriaceae) from Iraq were isolated and characterized with phylogenetic analysis based on the internal transcribed spacer (ITS) region. The efficacy of C. rosea against M. persicae and A. fabae was previously unknown. RESULTS: In the laboratory bioassays, mortality of adult M. persicae and A. fabae caused by both C. rosea isolates varied according to conidial concentrations, with complete mortality occurring at 1 × 109 conidia ml-1 10 day post treatment. For M. persicae, LC50 values of AA80 and AA80 isolates were 3.6 × 106 and 3.8 × 106 conidia ml-1. For A. fabae, LC50 values of AA80 and AA80 isolates were 4.5 × 106 and 4.35 × 106 conidia ml-1. Infection by both fungal isolates at LC50 values reduced total fecundity of the treated aphids by 20% when compared to the untreated aphids. CONCLUSIONS: The results from laboratory bioassays showed that C. rosea has potential as a biological control agent of M. persicae and A. fabae which is crucial for ecofriendly biopesticide development. However, further field and greenhouse studies are required for mass production.

Insulin Sensitivity After Living Donor Nephrectomy.

BACKGROUND: The kidney is essential for glucose and insulin metabolism. Living kidney donors (LKDs) experience a reduction in glomerular filtration rate of 25 to 30 mL/min after donor nephrectomy. Little is known about the effect of glomerular filtration rate decline on insulin sensitivity in LKDs. METHODS: We conducted a prospective pilot study on 9 LKDs (N = 9) who underwent dynamic metabolic testing (mixed meal tolerance test) to measure proxies of insulin sensitivity (homeostatic model assessment of insulin resistance, the area under curve [AUC] for insulin/glucose ratio, and Matsuda insulin sensitivity index) before and 3 months after donor nephrectomy. The primary outcome was the change in insulin sensitivity indices (delta [post-nephrectomy - pre-nephrectomy]). RESULTS: Four of the donors had a body mass index (BMI) between 32.0 and 36.7 predonation. Post-donor nephrectomy, compared with prenephrectomy values, median insulin AUC increased from 60.7 to 101.7 hr*mU/mL (delta median 33.3, P = .04) without significant change in median glucose AUC levels from 228.9 to 209.3 hr*mg/dL (delta median 3.2, P = .77). There was an increase in the median homeostatic model assessment of insulin resistance from 2 to 2.9 (delta median 0.8, P = .03) and the AUC insulin/glucose ratio from 30.9 to 62.1 pmol/mmol (delta median 17.5, P = .001), whereas the median Matsuda insulin sensitivity index decreased from 5.9 to 2.9 (delta median -2, P = .05). The changes were more pronounced in obese (BMI >32) donors. CONCLUSION: LKDs appear to have a trend toward a decline in insulin sensitivity post-donor nephrectomy in the short term, especially in obese donors (BMI >32). Further investigation with a larger sample size and longer follow-up is needed.

Uncontrolled Confounders May Lead to False or Overvalued Radiomics Signature: A Proof of Concept Using Survival Analysis in a Multicenter Cohort of Kidney Cancer.

PURPOSE: We aimed to explore potential confounders of prognostic radiomics signature predicting survival outcomes in clear cell renal cell carcinoma (ccRCC) patients and demonstrate how to control for them. MATERIALS AND METHODS: Preoperative contrast enhanced abdominal CT scan of ccRCC patients along with pathological grade/stage, gene mutation status, and survival outcomes were retrieved from The Cancer Imaging Archive (TCIA)/The Cancer Genome Atlas-Kidney Renal Clear Cell Carcinoma (TCGA-KIRC) database, a publicly available dataset. A semi-automatic segmentation method was applied to segment ccRCC tumors, and 1,160 radiomics features were extracted from each segmented tumor on the CT images. Non-parametric principal component decomposition (PCD) and unsupervised hierarchical clustering were applied to build the radiomics signature models. The factors confounding the radiomics signature were investigated and controlled sequentially. Kaplan-Meier curves and Cox regression analyses were performed to test the association between radiomics signatures and survival outcomes. RESULTS: 183 patients of TCGA-KIRC cohort with available imaging, pathological, and clinical outcomes were included in this study. All 1,160 radiomics features were included in the first radiomics signature. Three additional radiomics signatures were then modelled in successive steps removing redundant radiomics features first, removing radiomics features biased by CT slice thickness second, and removing radiomics features dependent on tumor size third. The final radiomics signature model was the most parsimonious, unbiased by CT slice thickness, and independent of tumor size. This final radiomics signature stratified the cohort into radiomics phenotypes that are different by cancer-specific and recurrence-free survival; HR (95% CI) = 3.0 (1.5-5.7), p <0.05 and HR (95% CI) = 6.6 (3.1-14.1), p <0.05, respectively. CONCLUSION: Radiomics signature can be confounded by multiple factors, including feature redundancy, image acquisition parameters like slice thickness, and tumor size. Attention to and proper control for these potential confounders are necessary for a reliable and clinically valuable radiomics signature.

Prolonged Response to HER2-Directed Therapy in Three Patients with HER2-Amplified Metastatic Carcinoma of the Biliary System: Case Study and Review of the Literature.

HER2 amplification, which results in overexpression of the receptor tyrosine kinase HER2, has been described in a wide variety of malignancies. HER2-targeting agents have been incorporated into the treatment paradigms for HER2-overexpressing breast and gastric cancer. More recently, these agents have shown promise in other gastrointestinal malignancies, such as colon cancer and biliary tract tumors. This study discusses two patients with gallbladder carcinoma and a third with ampullary carcinoma who were able to achieve marked responses to HER2-directed therapy. These cases underscore the importance of molecular analysis for HER2 amplification/HER2 overexpression, irrespective of tumor histology, and highlight a need for further investigation of HER2-directed therapy beyond breast and gastroesophageal cancers. KEY POINTS: Current guidelines recommend molecular assessment for HER2 overexpression exclusively in breast and gastric adenocarcinoma. The focus of this report is on three cases (two biliary tract and one ampullary carcinoma) in which amplification of HER2 or overexpression of HER2 was detected and treatment with HER2-directed therapy resulted in robust responses. These cases exemplify responsiveness of non-breast/gastric histologies to HER2-directed therapies, highlighting several promising new settings for these agents. Testing for amplification of HER2 or overexpression of HER2 should be considered especially in rare diseases with limited treatment options.

The utility of an under-distended gallbladder on ultrasound in ruling out acute cholecystitis.

PURPOSE: To study the association between gallbladder dimensions and acute cholecystitis and to define a sensitive cutoff for excluding the disease. MATERIALS AND METHODS: 456 consecutive patients with an abdominal ultrasound performed for right upper quadrant pain, from 1/2019 to 4/2019, were retrospectively collected. Length and width of the gallbladder were measured by a blinded radiology fellow. Patient charts were examined for prospective sonographic findings, clinical data, and pathology from subsequent cholecystectomy or at least 1 month of follow-up with symptom resolution. Univariable and multivariable logistic regression analysis were conducted to define the association of gallbladder dimensions and other sonographic and clinical variables with acute cholecystitis. Optimal and sensitive cutoffs of gallbladder widths were defined. The determined sensitive cutoff was validated with a separate cohort of 501 consecutive patients. RESULTS: 319 patients (median age 48 ± 19 years) including 11%, 19%, and 70% with acute, chronic, and no cholecystitis were included in the experimental cohort, respectively, and 298 patients (median age 50 ± 19 years) including 10%, 12%, and 79% with acute, chronic, and no cholecystitis were included in the validation cohort, respectively. Of all sonographic findings and clinical data, gallbladder width produced the best univariate ROC curve with an AUC of 0.91 (Odds ratio 5.1, 95% CI 3.1-8.5, p < 0.001). 2.2 cm was the gallbladder width cutoff below which there were no cases of acute cholecystitis in the experimental cohort. Multivariable logistic regression analysis using sonographic findings only produced an ROC curve with an AUC of 0.94. Applying the 2.2 cm cutoff in the validation cohort resulted in 100% sensitivity. CONCLUSION: Lack of gallbladder distention, defined as a width less than 2.2 cm, has potential to serve as a highly sensitive sign for exclusion of acute cholecystitis, regardless of additional sonographic findings and clinical data.

Provision of a DAPT Score to Cardiologists and Extension of Dual Antiplatelet Therapy Beyond 1 Year After ACS: Randomized Substudy of the Prospective Canadian ACS Reflective II Study.

BACKGROUND: Extension of dual antiplatelet therapy (DAPT) beyond 1 year after acute coronary syndrome is associated with a reduction in ischemic events but also increased bleeding. The DAPT score identifies individuals likely to derive overall benefit or harm from DAPT extension. We sought to evaluate the impact of providing the DAPT score to treating physicians on the decision to extend DAPT beyond 1 year after non-ST-segment elevation myocardial infarction. METHODS: Moderate to high-risk non-ST-segment elevation myocardial infarction patients were enrolled from July 2016 to May 2018 in 13 Canadian hospitals by 52 cardiologists. Participating cardiologists were randomly assigned 1:1 to receive their individual patients' DAPT scores before the 1-year follow-up visit vs not receiving their patients' DAPT scores. Rates of DAPT extension were compared among the randomized groups. RESULTS: At 1 year, 370 of the 585 (63.2%) patients discharged on DAPT were receiving DAPT. Among patients on DAPT at 1 year, the median (25th, 75th percentile) DAPT score was 2 (1,3). DAPT was extended beyond 1 year in 36.2% randomly assigned to provision of DAPT score vs 35.7% in the control group (P = 0.93). In the subgroup of patients with DAPT score ≥ 2, DAPT extension was 49.5% in the DAPT score provision arm vs 40.4% in the control arm (P = 0.22); among patients with DAPT score < 2, DAPT termination was 78.6% in the DAPT score provision arm vs 70.6% in the control arm (P = 0.26) (P value for interaction = 0.1). CONCLUSIONS: In this exploratory randomized trial, provision of the DAPT score to treating physicians had no impact on the duration of DAPT treatment beyond 1 year.

INTRODUCTION: La prolongation de la bithérapie antiplaquettaire au-delà d'un an après un syndrome coronarien aigu est associée à la réduction des accidents ischémiques, mais aussi à l'augmentation des hémorragies. Le score de bithérapie antiplaquettaire permet de déterminer les individus susceptibles d'obtenir des avantages globaux ou des inconvénients de la prolongation de la bithérapie antiplaquettaire. Nous avons cherché à évaluer les répercussions de l'obtention du score de bithérapie antiplaquettaire par les médecins traitants sur la décision quant à la prolongation de la bithérapie antiplaquettaire au-delà d'un an après l'infarctus du myocarde sans élévation du segment ST. MÉTHODES: De juillet 2016 à mai 2018, 52 cardiologues de 13 hôpitaux du Canada ont inscrit des patients exposés à un risque modéré à élevé d'infarctus du myocarde sans élévation du segment ST. Nous avons réparti de façon aléatoire selon un rapport 1:1 les cardiologues participants qui recevaient les scores de bithérapie antiplaquettaire individuels de leurs patients avant la consultation de suivi après un an vs ceux qui ne recevaient pas les scores de bithérapie antiplaquettaire de leurs patients. Nous avons comparé les taux de prolongation de la bithérapie antiplaquettaire des groupes répartis de façon aléatoire. RÉSULTATS: Après un an, 370 (63,2 %) patients sur 585 qui avaient eu à la sortie de l'hôpital une bithérapie antiplaquettaire recevaient la bithérapie antiplaquettaire. Parmi les patients qui prenaient la bithérapie antiplaquettaire après un an, le score médian de bithérapie antiplaquettaire (25e, 75e percentiles) était de 2 (1, 3). La bithérapie antiplaquettaire était prolongée au-delà d'un an chez 36,2 % des patients répartis de façon aléatoire qui avaient un score de bithérapie antiplaquettaire vs 35,7 % dans le groupe témoin (P = 0,93). Dans le sous-groupe de patients qui avaient un score de bithérapie antiplaquettaire ≥ 2, la prolongation de la bithérapie antiplaquettaire était de 49,5 % dans le bras qui avait un score de bithérapie antiplaquettaire vs 40,4 % dans le bras témoin (P = 0,22); parmi les patients qui avaient un score de bithérapie antiplaquettaire < 2, la cessation de la bithérapie antiplaquettaire était de 78,6 % dans le bras qui avait un score de bithérapie antiplaquettaire vs 70,6 % dans le bras témoin (P = 0,26) (valeur P pour l'interaction = 0,1). CONCLUSIONS: Dans cet essai exploratoire à répartition aléatoire, l'obtention du score de la bithérapie antiplaquettaire par les médecins traitants n'a pas engendré de répercussions sur la durée de la bithérapie antiplaquettaire au-delà d'un an.

How Many Radiographs Does It Take to Screen for Femoral Cam Morphology?: A Noninferiority Study.

PURPOSE: To compare a 2-view radiograph series (AP of the pelvis and 45° Dunn of the hip) with a 5-view radiograph series for sensitivity in identifying femoral cam morphology. MATERIALS AND METHODS: This is a retrospective review of consecutive patients with a 5-view radiograph series (AP pelvis and AP, 45° Dunn, frog lateral, and false profile of the affected hip) from 2016 to 2017. Three fellowship trained radiologists blindly and independently evaluated 2 views (AP pelvis and Dunn) for a femoral cam lesion, acetabular rim calcification, Tonnis grade, and important incidental findings. Two weeks later, the same assessment was made on all 5 views. A noninferiority test of the 2-view series vs the 5-view series for sensitivity in identifying femoral cam morphology was conducted. Individual reader sensitivity calculations were performed and agreement was determined with the kappa statistic. RESULTS: The 2-view series was noninferior to the 5-view series for cam identification (P value = 0.010). In comparing the 2-view vs 5-view series for individual readers, there was no difference in the sensitivities (84%-100% vs 85%-98%, P = 0.85-1.0) or specificities (11%-56% vs 7%-56%, P = 0.58-1.0) for cam identification. There was fair to excellent 2-view intrareader agreement (k = 0.38-0.93) and similar inter-reader agreement between the 2-view and 5-view (k = 0.33 vs 0.37). CONCLUSIONS: A 2-view radiograph series (AP pelvis and Dunn hip) is noninferior to a 5-view radiograph series for sensitivity in identifying femoral cam morphology.

Comparing RECIST 1.1 and iRECIST in advanced melanoma patients treated with pembrolizumab in a phase II clinical trial.

OBJECTIVES: To compare tumor best overall response (BOR) by RECIST 1.1 and iRECIST, to explore the incidence of pseudoprogression in melanoma treated with pembrolizumab, and to assess the impact of pseudoprogression on overall survival (OS). METHODS: A total of 221 patients with locally advanced/unresectable melanoma who received pembrolizumab as part of KEYNOTE-002 trial were included in this study. Radiological assessment of imaging was centrally reviewed to assess tumor response. Incidence of discordance in BOR between RECIST 1.1 and iRECIST as well as rate of pseudoprogression were measured. OS of patients with pseudoprogression was compared with that of those with uncontrolled disease. RESULTS: Of the 221 patients in this cohort, 136 patients developed PD as per RECIST v1.1 and 78 patients with PD continued treatment and imaging beyond initial RECIST 1.1-defined PD. Among the 78 patients who continued therapy and imaging post-progression, RECIST 1.1 and iRECIST were discordant in 10 patients (12.8%) and pseudoprogression was encountered in 14 patients (17.9%). OS of patients with pseudoprogression was longer than that of patients with uncontrolled disease/true progression (29.9 months versus 8.0 months, p value < 0.001). CONCLUSIONS: Effectiveness of immunotherapy in clinical trials depends on the criterion used to assess tumor response (RECIST 1.1 vs iRECIST) with iRECIST being more appropriate to detect pseudoprogression and potentially prevent premature termination of effective therapy. Pseudoprogression was associated with improved OS in comparison with that of patients with uncontrolled disease. KEY POINTS: • Discordance between iRECIST and RECIST 1.1 was found in 12.8% of unresectable melanoma patients on pembrolizumab who continued therapy beyond initial RECIST 1.1-defined progression. • Pseudoprogression, captured with iRECIST, occurred in 17.9% and was significantly associated with improved overall survival in comparison with uncontrolled disease.

CT colonography's role in the COVID-19 pandemic: a safe(r), socially distanced total colon examination.

PURPOSE: To describe the favorable procedural profile of CT colonography (CTC) during the COVID-19 pandemic. CONCLUSION: Postponement of cancer screening due to COVID-19 has resulted in a backlog of individuals needing to undergo structural examination of the colon. The experience during the initial COVID-19 surge with urgent evaluation of the colon for transplant patients prior to transplant suggests that CTC can be done in a lower risk manner as compared to other structural examinations. The procedural profile of CTC is advantageous during this pandemic as maintaining social distancing and preserving healthcare supplies including PPE are of paramount importance. CTC is an important option to utilize in the screening armamentarium to allow effective screening of average risk asymptomatic individuals in the COVID-19 era.

Deep Learning of Computed Tomography Virtual Wedge Resection for Prediction of Histologic Usual Interstitial Pneumonitis.

Rationale: The computed tomography (CT) pattern of definite or probable usual interstitial pneumonia (UIP) can be diagnostic of idiopathic pulmonary fibrosis and may obviate the need for invasive surgical biopsy. Few machine-learning studies have investigated the classification of interstitial lung disease (ILD) on CT imaging, but none have used histopathology as a reference standard.Objectives: To predict histopathologic UIP using deep learning of high-resolution computed tomography (HRCT).Methods: Institutional databases were retrospectively searched for consecutive patients with ILD, HRCT, and diagnostic histopathology from 2011 to 2014 (training cohort) and from 2016 to 2017 (testing cohort). A blinded expert radiologist and pulmonologist reviewed all training HRCT scans in consensus and classified HRCT scans based on the 2018 American Thoracic Society/European Respriatory Society/Japanese Respiratory Society/Latin American Thoracic Association diagnostic criteria for idiopathic pulmonary fibrosis. A convolutional neural network (CNN) was built accepting 4 × 4 × 2 cm virtual wedges of peripheral lung on HRCT as input and outputting the UIP histopathologic pattern. The CNN was trained and evaluated on the training cohort using fivefold cross validation and was then tested on the hold-out testing cohort. CNN and human performance were compared in the training cohort. Logistic regression and survival analyses were performed.Results: The CNN was trained on 221 patients (median age 60 yr; interquartile range [IQR], 53-66), including 71 patients (32%) with UIP or probable UIP histopathologic patterns. The CNN was tested on a separate hold-out cohort of 80 patients (median age 66 yr; IQR, 58-69), including 22 patients (27%) with UIP or probable UIP histopathologic patterns. An average of 516 wedges were generated per patient. The percentage of wedges with CNN-predicted UIP yielded a cross validation area under the curve of 74% for histopathological UIP pattern per patient. The optimal cutoff point for classifying patients on the training cohort was 16.5% of virtual lung wedges with CNN-predicted UIP and resulted in sensitivity and specificity of 74% and 58%, respectively, in the testing cohort. CNN-predicted UIP was associated with an increased risk of death or lung transplantation during cross validation (hazard ratio, 1.5; 95% confidence interval, 1.1-2.2; P = 0.03).Conclusions: Virtual lung wedge resection in patients with ILD can be used as an input to a CNN for predicting the histopathologic UIP pattern and transplant-free survival.

Impact of mandated prospectively reported apparent diffusion coefficient values on the rates of positivity for clinically significant prostate cancer by PI-RADS score.

BACKGROUND: Prior research has shown that retrospectively measured apparent diffusion coefficient (ADC) of prostate magnetic resonance imaging (MRI) lesions is associated with clinically significant prostate cancer (csPCa) on targeted biopsy suggesting that ADC should be measured and reported prospectively. PURPOSE: To assess the impact of mandatory prospective measurement of ADC on the rates of positivity across PI-RADS scores for csPCa. MATERIAL AND METHODS: Consecutive patients who underwent ultrasound (US)-MRI fusion prostate biopsy from August 2018 to July 2019 and who had prospectively reported ADC were compared to control patients who did not. Rates of positivity by PI-RADS category were computed and compared using Chi-square. Multivariable regression was performed. RESULTS: In total, 126 patients (median age 65 years) with 165 prostate lesions (19, 51, 70, and 25 PI-RADS 2, 3, 4, and 5, respectively) and prospectively reported ADC values were compared to 113 control patients (median age 66 years) with 157 prostate lesions (17, 42, 64, and 34 PI-RADS 2, 3, 4, and 5, respectively). Rates of positivity across PI-RADS scores were similar between the two cohorts; 11%, 25%, 55%, and 76% and 0%, 21%, 56%, and 62% for PI-RADS 2, 3, 4, and 5 in the test and control cohorts, respectively (Chi-square P = 0.78). Multivariate logistic regression showed no significant association between the presence of prospectively measured ADC and csPCa (odds ratio 1.1, 95% confidence interval 0.7-1.7, P = 0.82). CONCLUSION: Prospective ADC measurement may not impact PI-RADS category assignments or positivity rates for csPCa under current guidelines. Future versions of PI-RADS may need to incorporate ADC into scoring rules to realize their potential.

Enhanced Detection of Treatment Effects on Metastatic Colorectal Cancer with Volumetric CT Measurements for Tumor Burden Growth Rate Evaluation.

PURPOSE: Mathematical models combined with new imaging technologies could improve clinical oncology studies. To improve detection of therapeutic effect in patients with cancer, we assessed volumetric measurement of target lesions to estimate the rates of exponential tumor growth and regression as treatment is administered. EXPERIMENTAL DESIGN: Two completed phase III trials were studied (988 patients) of aflibercept or panitumumab added to standard chemotherapy for advanced colorectal cancer. Retrospectively, radiologists performed semiautomated measurements of all metastatic lesions on CT images. Using exponential growth modeling, tumor regression (d) and growth (g) rates were estimated for each patient's unidimensional and volumetric measurements. RESULTS: Exponential growth modeling of volumetric measurements detected different empiric mechanisms of effect for each drug: panitumumab marginally augmented the decay rate [tumor half-life; d [IQR]: 36.5 days (56.3, 29.0)] of chemotherapy [d: 44.5 days (67.2, 32.1), two-sided Wilcoxon P = 0.016], whereas aflibercept more significantly slowed the growth rate [doubling time; g = 300.8 days (154.0, 572.3)] compared with chemotherapy alone [g = 155.9 days (82.2, 347.0), P ≤ 0.0001]. An association of g with overall survival (OS) was observed. Simulating clinical trials using volumetric or unidimensional tumor measurements, fewer patients were required to detect a treatment effect using a volumetric measurement-based strategy (32-60 patients) than for unidimensional measurement-based strategies (124-184 patients). CONCLUSIONS: Combined tumor volume measurement and estimation of tumor regression and growth rate has potential to enhance assessment of treatment effects in clinical studies of colorectal cancer that would not be achieved with conventional, RECIST-based unidimensional measurements.

The Negative Predictive Value of a PI-RADS Version 2 Score of 1 on Prostate MRI and the Factors Associated With a False-Negative MRI Study.

OBJECTIVE. The purpose of this study was to calculate the negative predictive value of a prostate MRI study with a Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) score of 1 (hereafter referred to as a PI-RADS 1 MRI study) and to explore the patient characteristics and MRI-based factors associated with an MRI study with false-negative results. MATERIALS AND METHODS. A total of 542 consecutive patients with a PI-RADS 1 MRI study obtained between January 2016 and July 2019 were retrospectively identified. Patient charts were examined to identify those patients who subsequently underwent systematic prostate biopsy within 1 year of undergoing MRI or at any later date if the biopsy was negative. Patient characteristics and MRI-specific factors were recorded. Two blinded radiologists evaluated the quality of the axial T2-weighted, DWI, and apparent diffusion coefficient sequences; measured the volume of the bladder, the prostate gland, and rectal gas; and determined whether the peripheral zone was avidly enhancing and whether low signal intensity was seen in 50% or more of the peripheral zone on T2-weighted images. Interobserver agreement was tested. Univariable and multivariable logistic regression models were built. RESULTS. A total of 150 patients (median age, 63 years; interquartile range, 56-70 years) were included. Of these patients, 19 (13%) had prostate cancer with a Gleason score of 3 + 4 or greater, yielding a negative predictive value of 87%. Both low T2 signal intensity in the peripheral zone and the prostate-specific antigen level were associated with a false-negative PI-RADS 1 assessment (odds ratio, 4.9 [95% CI, 1.6-14.9; p = 0.006] and 1.1 [95% CI, 1.0-1.2; p = 0.03], respectively). A cutoff prostate-specific antigen level of 3.97 ng/mL resulted in sensitivity and specificity of 89% and 21%, respectively. There was moderate interobserver agreement for low T2 signal intensity in the peripheral zone (κ coefficient = 0.75). CONCLUSION. Even among select patients who undergo subsequent biopsy because of a high clinical suspicion of prostate cancer, a PI-RADS 1 prostate MRI study has a high negative predictive value. A T2-hypointense peripheral zone and an elevated prostate-specific antigen level are significantly associated with a false-negative MRI study.