BackgroundThere is interest in using convolutional neural networks (CNNs) to analyze medical imaging to provide computer-aided diagnosis (CAD). Recent work has suggested that image classification CNNs may not generalize to new data as well as previously believed. We assessed how well CNNs generalized across three hospital systems for a simulated pneumonia screening task.Methods and findingsA cross-sectional design with multiple model training cohorts was used to evaluate model generalizability to external sites using split-sample validation. A total of 158,323 chest radiographs were drawn from three institutions: National Institutes of Health Clinical Center (NIH; 112,120 from 30,805 patients), Mount Sinai Hospital (MSH; 42,396 from 12,904 patients), and Indiana University Network for Patient Care (IU; 3,807 from 3,683 patients). These patient populations had an age mean (SD) of 46.9 years (16.6), 63.2 years (16.5), and 49.6 years (17) with a female percentage of 43.5%, 44.8%, and 57.3%, respectively. We assessed individual models using the area under the receiver operating characteristic curve (AUC) for radiographic findings consistent with pneumonia and compared performance on different test sets with DeLong’s test. The prevalence of pneumonia was high enough at MSH (34.2%) relative to NIH and IU (1.2% and 1.0%) that merely sorting by hospital system achieved an AUC of 0.861 (95% CI 0.855–0.866) on the joint MSH–NIH dataset. Models trained on data from either NIH or MSH had equivalent performance on IU (P values 0.580 and 0.273, respectively) and inferior performance on data from each other relative to an internal test set (i.e., new data from within the hospital system used for training data; P values both <0.001). The highest internal performance was achieved by combining training and test data from MSH and NIH (AUC 0.931, 95% CI 0.927–0.936), but this model demonstrated significantly lower external performance at IU (AUC 0.815, 95% CI 0.745–0.885, P = 0.001). To test the effect of pooling data from sites with disparate pneumonia prevalence, we used stratified subsampling to generate MSH–NIH cohorts that only differed in disease prevalence between training data sites. When both training data sites had the same pneumonia prevalence, the model performed consistently on external IU data (P = 0.88). When a 10-fold difference in pneumonia rate was introduced between sites, internal test performance improved compared to the balanced model (10× MSH risk P < 0.001; 10× NIH P = 0.002), but this outperformance failed to generalize to IU (MSH 10× P < 0.001; NIH 10× P = 0.027). CNNs were able to directly detect hospital system of a radiograph for 99.95% NIH (22,050/22,062) and 99.98% MSH (8,386/8,388) radiographs. The primary limitation of our approach and the available public data is that we cannot fully assess what other factors might be contributing to hospital system–specific biases.ConclusionPneumonia-screening CNNs achieved better internal than external performance in 3 out of 5 natural comparisons. When models wer...

Activation of cyclin-dependent kinases 4 and 6 (cdk4/6) occurs in the majority of glioblastoma multiforme (GBM) tumors, and represents a promising molecular target for the development of small molecule inhibitors. In the current study, we investigated the molecular determinants and in vivo response of diverse GBM cell lines and xenografts to PD-0332991, a cdk4/6-specific inhibitor. In vitro testing of PD-0332991 against a panel of GBM cell lines revealed a potent G 1 cell cycle arrest and induction of senescence in each of 16 retinoblastoma protein (Rb)-proficient cell lines regardless of other genetic lesions, whereas 5 cell lines with homozy-

BackgroundStereotactic body radiation therapy (SBRT) delivers fewer high-dose fractions of radiation which may be radiobiologically favorable to conventional low-dose fractions commonly used for prostate cancer radiotherapy. We report our early experience using SBRT for localized prostate cancer.MethodsPatients treated with SBRT from June 2008 to May 2010 at Georgetown University Hospital for localized prostate carcinoma, with or without the use of androgen deprivation therapy (ADT), were included in this retrospective review of data that was prospectively collected in an institutional database. Treatment was delivered using the CyberKnife® with doses of 35 Gy or 36.25 Gy in 5 fractions. Biochemical control was assessed using the Phoenix definition. Toxicities were recorded and scored using the CTCAE v.3. Quality of life was assessed before and after treatment using the Short Form-12 Health Survey (SF-12), the American Urological Association Symptom Score (AUA) and Sexual Health Inventory for Men (SHIM) questionnaires. Late urinary symptom flare was defined as an AUA score ≥ 15 with an increase of ≥ 5 points above baseline six months after the completion of SBRT.ResultsOne hundred patients (37 low-, 55 intermediate- and 8 high-risk according to the D’Amico classification) at a median age of 69 years (range, 48–90 years) received SBRT, with 11 patients receiving ADT. The median pre-treatment prostate-specific antigen (PSA) was 6.2 ng/ml (range, 1.9-31.6 ng/ml) and the median follow-up was 2.3 years (range, 1.4-3.5 years). At 2 years, median PSA decreased to 0.49 ng/ml (range, 0.1-1.9 ng/ml). Benign PSA bounce occurred in 31% of patients. There was one biochemical failure in a high-risk patient, yielding a two-year actuarial biochemical relapse free survival of 99%. The 2-year actuarial incidence rates of GI and GU toxicity ≥ grade 2 were 1% and 31%, respectively. A median baseline AUA symptom score of 8 significantly increased to 11 at 1 month (p = 0.001), however returned to baseline at 3 months (p = 0.60). Twenty one percent of patients experienced a late transient urinary symptom flare in the first two years following treatment. Of patients who were sexually potent prior to treatment, 79% maintained potency at 2 years post-treatment.ConclusionsSBRT for clinically localized prostate cancer was well tolerated, with an early biochemical response similar to other radiation therapy treatments. Benign PSA bounces were common. Late GI and GU toxicity rates were comparable to conventionally fractionated radiation therapy and brachytherapy. Late urinary symptom flares were observed but the majority resolved with conservative management. A high percentage of men who were potent prior to treatment remained potent two years following treatment.

Rapid diagnosis and treatment of acute neurological illnesses such as stroke, hemorrhage, and hydrocephalus are critical to achieving positive outcomes and preserving neurologic function-'time is brain'. Although these disorders are often recognizable by their symptoms, the critical means of their diagnosis is rapid imaging. Computer-aided surveillance of acute neurologic events in cranial imaging has the potential to triage radiology workflow, thus decreasing time to treatment and improving outcomes. Substantial clinical work has focused on computer-assisted diagnosis (CAD), whereas technical work in volumetric image analysis has focused primarily on segmentation. 3D convolutional neural networks (3D-CNNs) have primarily been used for supervised classification on 3D modeling and light detection and ranging (LiDAR) data. Here, we demonstrate a 3D-CNN architecture that performs weakly supervised classification to screen head CT images for acute neurologic events. Features were automatically learned from a clinical radiology dataset comprising 37,236 head CTs and were annotated with a semisupervised natural-language processing (NLP) framework. We demonstrate the effectiveness of our approach to triage radiology workflow and accelerate the time to diagnosis from minutes to seconds through a randomized, double-blinded, prospective trial in a simulated clinical environment.

We propose the MRRC to further differentiate Class III aneurysms into those likely to progress to complete occlusion and those likely to remain incompletely occluded or to worsen. The MRRC has the potential to expand the definition of adequate coil embolization, possibly decrease procedural risk, and help endovascular neurosurgeons predict which patients need closer angiographic follow-up. These findings need to be validated in a prospective study with independent blinded angiographic grading.

T he scientific, academic, medical and data science communities have come together in the face of the COVID-19 pandemic crisis to rapidly assess novel paradigms in artificial intelligence (AI) that are rapid and secure, and potentially incentivize data sharing and model training and testing without the usual privacy and data ownership hurdles of conventional collaborations 1,2 . Healthcare providers, researchers and industry have pivoted their focus to address unmet and critical clinical needs created by the crisis, with remarkable results [3][4][5][6][7][8][9] . Clinical trial recruitment has been expedited and facilitated by national regulatory bodies and an international cooperative spirit 10-12 . The data analytics and AI disciplines have always fostered open

Aneurysmal subarachnoid hemorrhage (SAH) affects six to nine people per 100,000 per year, has a 35% mortality, and leaves many with lasting disabilities, often related to cognitive dysfunction. Clinical decision rules and more sensitive computed tomography (CT) have made the diagnosis of SAH easier, but physicians must maintain a high index of suspicion. The management of these patients is based on a limited number of randomized clinical trials (RCTs). Early repair of the ruptured aneurysm by endovascular coiling or neurosurgical clipping is essential, and coiling is superior to clipping in cases amenable to both treatments. Aneurysm repair prevents rebleeding, leaving the most important prognostic factors for outcome early brain injury from the hemorrhage, which is reflected in the neurologic condition of the patient, and delayed cerebral ischemia (DCI). Observational studies suggest outcomes are better when patients are managed in specialized neurologic intensive care units with inter-or multidisciplinary clinical groups. Medical management aims to minimize early brain injury, cerebral edema, hydrocephalus, increased intracranial pressure (ICP), and medical complications. Management then focuses on preventing, detecting, and treating DCI. Nimodipine is the only pharmacologic treatment that is approved for SAH in most countries, as no other intervention has demonstrated efficacy. In fact, much of SAH management is derived from studies in other patient populations. Therefore, further study of complications, including DCI and other medical complications, is needed to optimize outcomes for this fragile patient population.

Prostate cancer (PCa) is the most common type of cancer in men in the United States, which disproportionately affects African American descents. While metastasis is the most common cause of death among PCa patients, no specific markers have been assigned to severity and ethnic biasness of the disease. MicroRNAs represent a promising new class of biomarkers owing to their inherent stability and resilience. In the present study, we investigated potential miRNAs that can be used as biomarkers and/or therapeutic targets and can provide insight into the severity and ethnic biasness of PCa. PCR array was performed in FFPE PCa tissues (5 Caucasian American and 5 African American) and selected differentially expressed miRNAs were validated by qRT-PCR, in 40 (15 CA and 25 AA) paired PCa and adjacent normal tissues. Significantly deregulated miRNAs were also analyzed in urine samples to explore their potential as non-invasive biomarker for PCa. Out of 8 miRNAs selected for validation from PCR array data, miR-205 (p<0.0001), mir-214 (p<0.0001), miR-221(p<0.001) and miR-99b (p<0.0001) were significantly downregulated in PCa tissues. ROC curve shows that all four miRNAs successfully discriminated between PCa and adjacent normal tissues. MiR-99b showed significant down regulation (p<0.01) in AA PCa tissues as compared to CA PCa tissues and might be related to the aggressiveness associated with AA population. In urine, miR-205 (p<0.05) and miR-214 (p<0.05) were significantly downregulated in PCa patients and can discriminate PCa patients from healthy individuals with 89% sensitivity and 80% specificity. In conclusion, present study showed that miR-205 and miR-214 are downregulated in PCa and may serve as potential non-invasive molecular biomarker for PCa.