RAPHIA: A deep learning pipeline for the registration of MRI and whole-mount histopathology images of the prostate.

Image registration can map the ground truth extent of prostate cancer from histopathology images onto MRI, facilitating the development of machine learning methods for early prostate cancer detection. Here, we present RAdiology PatHology Image Alignment (RAPHIA), an end-to-end pipeline for efficient and accurate registration of MRI and histopathology images. RAPHIA automates several time-consuming manual steps in existing approaches including prostate segmentation, estimation of the rotation angle and horizontal flipping in histopathology images, and estimation of MRI-histopathology slice correspondences. By utilizing deep learning registration networks, RAPHIA substantially reduces computational time. Furthermore, RAPHIA obviates the need for a multimodal image similarity metric by transferring histopathology image representations to MRI image representations and vice versa. With the assistance of RAPHIA, novice users achieved expert-level performance, and their mean error in estimating histopathology rotation angle was reduced by 51% (12 degrees vs 8 degrees), their mean accuracy of estimating histopathology flipping was increased by 5% (95.3% vs 100%), and their mean error in estimating MRI-histopathology slice correspondences was reduced by 45% (1.12 slices vs 0.62 slices). When compared to a recent conventional registration approach and a deep learning registration approach, RAPHIA achieved better mapping of histopathology cancer labels, with an improved mean Dice coefficient of cancer regions outlined on MRI and the deformed histopathology (0.44 vs 0.48 vs 0.50), and a reduced mean per-case processing time (51 vs 11 vs 4.5 min). The improved performance by RAPHIA allows efficient processing of large datasets for the development of machine learning models for prostate cancer detection on MRI. Our code is publicly available at: https://github.com/pimed/RAPHIA.

Solvation Plays a Key Role in Antioxidant-Mediated Attenuation of Elevated Creatinine Level: An In Vitro Spectroscopic Investigation.

An elevated level of creatinine (CRN) is a mark of kidney ailment, and prolonged retention of such condition could lead to renal failure, associated with severe ischemia. Antioxidants are clinically known to excrete CRN from the body through urine, thereby reducing its level in blood. The molecular mechanism of such an exclusion process is still illusive. As the excretion channel is urine, solvation of the solute is expected to play a pivotal role. Here, we report a detailed time-domain and frequency-domain terahertz (THz) spectroscopic investigation to understand the solvation of CRN in the presence of two model antioxidants, mostly used to treat elevated CRN level: N-Acetyl-l-cysteine (NAC) and ascorbic acid (ASC). FTIR spectroscopy in the mid-infrared region and UV absorption spectroscopy measurements coupled with quantum chemical calculations [at the B3LYP/6-311G++(d,p) level] reveal that both NAC and ASC form HBonded complexes with CRN and rapidly undergo a barrier-less proton transfer process to form creatinium ions. THz measurements provide explicit evidence of the formation of highly solvated complexes compared with bare CRN, which eventually enables its excretion through urine. These observations could provide a foundation for designing more beneficial drugs to resolve kidney diseases..

Safety profile of cyclin-dependent kinase (CDK) 4/6 inhibitors with concurrent radiation therapy: A systematic review and meta-analysis.

The cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6i) have become the standard of care for hormone receptor-positive (HR + ) and human epidermal growth factor receptor 2-negative (HER2-) metastatic breast cancer, improving survival outcomes compared to endocrine therapy alone. Abemaciclib and ribociclib, in combination with endocrine therapy, have demonstrated significant benefits in invasive disease-free survival for high-risk HR+/HER2- early breast cancer patients. Each CDK4/6i-palbociclib, ribociclib, and abemaciclib-exhibits distinct toxicity profiles. Radiation therapy (RT) can be delivered with a palliative or ablative intent, particularly using stereotactic body radiation therapy for oligometastatic or oligoprogressive disease. However, pivotal randomized trials lack information on concomitant CDK4/6i and RT, and existing preclinical and clinical data on the potential combined toxicities are limited and conflicting. As part of a broader effort to establish international consensus recommendations for integrating RT and targeted agents in breast cancer treatment, we conducted a systematic review and meta-analysis to evaluate the safety profile of combining CDK4/6i with palliative and ablative RT in both metastatic and early breast cancer settings.

Dose-escalated simultaneous integrated boost radiotherapy in early breast cancer (IMPORT HIGH): a multicentre, phase 3, non-inferiority, open-label, randomised controlled trial.

BACKGROUND: A tumour-bed boost delivered after whole-breast radiotherapy increases local cancer-control rates but requires more patient visits and can increase breast hardness. IMPORT HIGH tested simultaneous integrated boost against sequential boost with the aim of reducing treatment duration while maintaining excellent local control and similar or reduced toxicity. METHODS: IMPORT HIGH is a phase 3, non-inferiority, open-label, randomised controlled trial that recruited women after breast-conserving surgery for pT1-3pN0-3aM0 invasive carcinoma from radiotherapy and referral centres in the UK. Patients were randomly allocated to receive one of three treatments in a 1:1:1 ratio, with computer-generated random permuted blocks used to stratify patients by centre. The control group received 40 Gy in 15 fractions to the whole breast and 16 Gy in 8 fractions sequential photon tumour-bed boost. Test group 1 received 36 Gy in 15 fractions to the whole breast, 40 Gy in 15 fractions to the partial breast, and 48 Gy in 15 fractions concomitant photon boost to the tumour-bed volume. Test group 2 received 36 Gy in 15 fractions to the whole breast, 40 Gy in 15 fractions to the partial breast, and 53 Gy in 15 fractions concomitant photon boost to the tumour-bed volume. The boost clinical target volume was the clip-defined tumour bed. Patients and clinicians were not masked to treatment allocation. The primary endpoint was ipsilateral breast tumour relapse (IBTR) analysed by intention to treat; assuming 5% 5-year incidence with the control group, non-inferiority was predefined as 3% or less absolute excess in the test groups (upper limit of two-sided 95% CI). Adverse events were assessed by clinicians, patients, and photographs. This trial is registered with the ISRCTN registry, ISRCTN47437448, and is closed to new participants. FINDINGS: Between March 4, 2009, and Sept 16, 2015, 2617 patients were recruited. 871 individuals were assigned to the control group, 874 to test group 1, and 872 to test group 2. Median boost clinical target volume was 13 cm3 (IQR 7 to 22). At a median follow-up of 74 months there were 76 IBTR events (20 for the control group, 21 for test group 1, and 35 for test group 2). 5-year IBTR incidence was 1·9% (95% CI 1·2 to 3·1) for the control group, 2·0% (1·2 to 3·2) for test group 1, and 3·2% (2·2 to 4·7) for test group 2. The estimated absolute differences versus the control group were 0·1% (-0·8 to 1·7) for test group 1 and 1·4% (0·03 to 3·8) for test group 2. The upper confidence limit for test group 1 versus the control group indicated non-inferiority for 48 Gy. Cumulative 5-year incidence of clinician-reported moderate or marked breast induration was 11·5% for the control group, 10·6% for test group 1 (p=0·40 vs control group), and 15·5% for test group 2 (p=0·015 vs control group). INTERPRETATION: In all groups 5-year IBTR incidence was lower than the 5% originally expected regardless of boost sequencing. Dose-escalation is not advantageous. 5-year moderate or marked adverse event rates were low using small boost volumes. Simultaneous integrated boost in IMPORT HIGH was safe and reduced patient visits. FUNDING: Cancer Research UK.

Network pharmacology of apigeniflavan: a novel bioactive compound of Trema orientalis Linn. in the treatment of pancreatic cancer through bioinformatics approaches.

Pancreatic cancer is the seventh most prevalent cause of mortality globally. Since time immemorial, plant-derived products have been in use as therapeutic agents due to the existence of biologically active molecules called secondary metabolites. Flavonoids obtained from plants participate in cell cycle arrest, induce autophagy and apoptosis, and decrease oxidative stress in pancreatic cancer. The present study involves network pharmacology-based study of the methanolic leaf extract of Trema orientalis (MLETO) Linn. From the high-resolution mass spectrometry (HRMS) analysis, 21 nucleated flavonoids were screened out, of which only apigeniflavan was selected for further studies because it followed Lipinski's rule and showed no toxicity. The pharmacokinetics and physiochemical characteristics of apigeniflavan were performed using the online web servers pkCSM, Swiss ADME, and ProTox-II. This is the first in silico study to report the efficiency of apigeniflavan in pancreatic cancer treatment. The targets of apigeniflavan were fetched from SwissTargetPrediction database. The targets of pancreatic cancer were retrieved from DisGeNET and GeneCards. The protein-protein interaction of the common genes using Cytoscape yielded the top five hub genes: KDR, VEGFA, AKT1, SRC, and ESR1. Upon molecular docking, the lowest binding energies corresponded to best docking score which indicated the highest protein-ligand affinity. Kyoto Encyclopaedia of Genes and Genomes (KEGG) database was employed to see the involvement of hub genes in pathways related to pancreatic cancer. The following, pancreatic cancer pathway, MAPK, VEGF, PI3K-Akt, and ErbB signaling pathways, were found to be significant. Our results indicate the involvement of the hub genes in tumor growth, invasion and proliferation in the above-mentioned pathways, and therefore necessitating their downregulation. Moreover, apigeniflavan can flourish as a promising drug for the treatment of pancreatic cancer in future.

Network pharmacology-based anti-pancreatic cancer potential of kaempferol and catechin of Trema orientalis L. through computational approach.

In pancreatic cancer, healthy cells in the pancreas begin to malfunction and proliferate out of control. According to our conventional knowledge, many plants contain several novel bioactive compounds, having pharmaceutical applications for the treatment of disease like pancreatic cancer. The methanolic fraction of fruit extract of Trema orientalis L. (MFETO) was analysed through HRMS. In this in silico study, pharmacokinetic and physicochemical properties of the identified flavonoids from MFETO were screened out by ADMET analysis. Kaempferol and catechin followed Lipinski rules and showed no toxicity in Protox II. Targets of these compounds were taken from SwissTarget prediction and TCMSP whilst targets for pancreatic cancer were taken from GeneCards and DisGeNET databases. The protein-protein interaction (PPI) network of common genes was generated through STRING and then exported to the Cytoscape to get top 5 hub genes (AKT1, SRC, EGFR, TNF, and CASP3). The interaction between compounds and hub genes was analysed using molecular docking, and high binding affinity between them can be visualised by Biovia discovery studio visualizer. Our study shows that, five hub genes related to pancreatic cancer play an important role in tumour growth induction, invasion and migration. Kaempferol effectively check cell migration by inhibiting ERK1/2, EGFR-related SRC, and AKT pathways by scavenging ROS whilst catechin inhibited TNFα-induced activation and cell cycle arrest at G1 and G2/M phases by induction of apoptosis of malignant cells. Kaempferol and catechin containing MFETO can be used for formulation of potent drugs for pancreatic cancer treatment in future.

A review of artificial intelligence in prostate cancer detection on imaging.

A multitude of studies have explored the role of artificial intelligence (AI) in providing diagnostic support to radiologists, pathologists, and urologists in prostate cancer detection, risk-stratification, and management. This review provides a comprehensive overview of relevant literature regarding the use of AI models in (1) detecting prostate cancer on radiology images (magnetic resonance and ultrasound imaging), (2) detecting prostate cancer on histopathology images of prostate biopsy tissue, and (3) assisting in supporting tasks for prostate cancer detection (prostate gland segmentation, MRI-histopathology registration, MRI-ultrasound registration). We discuss both the potential of these AI models to assist in the clinical workflow of prostate cancer diagnosis, as well as the current limitations including variability in training data sets, algorithms, and evaluation criteria. We also discuss ongoing challenges and what is needed to bridge the gap between academic research on AI for prostate cancer and commercial solutions that improve routine clinical care.

Domain generalization for prostate segmentation in transrectal ultrasound images: A multi-center study.

Prostate biopsy and image-guided treatment procedures are often performed under the guidance of ultrasound fused with magnetic resonance images (MRI). Accurate image fusion relies on accurate segmentation of the prostate on ultrasound images. Yet, the reduced signal-to-noise ratio and artifacts (e.g., speckle and shadowing) in ultrasound images limit the performance of automated prostate segmentation techniques and generalizing these methods to new image domains is inherently difficult. In this study, we address these challenges by introducing a novel 2.5D deep neural network for prostate segmentation on ultrasound images. Our approach addresses the limitations of transfer learning and finetuning methods (i.e., drop in performance on the original training data when the model weights are updated) by combining a supervised domain adaptation technique and a knowledge distillation loss. The knowledge distillation loss allows the preservation of previously learned knowledge and reduces the performance drop after model finetuning on new datasets. Furthermore, our approach relies on an attention module that considers model feature positioning information to improve the segmentation accuracy. We trained our model on 764 subjects from one institution and finetuned our model using only ten subjects from subsequent institutions. We analyzed the performance of our method on three large datasets encompassing 2067 subjects from three different institutions. Our method achieved an average Dice Similarity Coefficient (Dice) of 94.0±0.03 and Hausdorff Distance (HD95) of 2.28 mm in an independent set of subjects from the first institution. Moreover, our model generalized well in the studies from the other two institutions (Dice: 91.0±0.03; HD95: 3.7 mm and Dice: 82.0±0.03; HD95: 7.1 mm). We introduced an approach that successfully segmented the prostate on ultrasound images in a multi-center study, suggesting its clinical potential to facilitate the accurate fusion of ultrasound and MRI images to drive biopsy and image-guided treatments.

Computational Detection of Extraprostatic Extension of Prostate Cancer on Multiparametric MRI Using Deep Learning.

The localization of extraprostatic extension (EPE), i.e., local spread of prostate cancer beyond the prostate capsular boundary, is important for risk stratification and surgical planning. However, the sensitivity of EPE detection by radiologists on MRI is low (57% on average). In this paper, we propose a method for computational detection of EPE on multiparametric MRI using deep learning. Ground truth labels of cancers and EPE were obtained in 123 patients (38 with EPE) by registering pre-surgical MRI with whole-mount digital histopathology images from radical prostatectomy. Our approach has two stages. First, we trained deep learning models using the MRI as input to generate cancer probability maps both inside and outside the prostate. Second, we built an image post-processing pipeline that generates predictions for EPE location based on the cancer probability maps and clinical knowledge. We used five-fold cross-validation to train our approach using data from 74 patients and tested it using data from an independent set of 49 patients. We compared two deep learning models for cancer detection: (i) UNet and (ii) the Correlated Signature Network for Indolent and Aggressive prostate cancer detection (CorrSigNIA). The best end-to-end model for EPE detection, which we call EPENet, was based on the CorrSigNIA cancer detection model. EPENet was successful at detecting cancers with extraprostatic extension, achieving a mean area under the receiver operator characteristic curve of 0.72 at the patient-level. On the test set, EPENet had 80.0% sensitivity and 28.2% specificity at the patient-level compared to 50.0% sensitivity and 76.9% specificity for the radiologists. To account for spatial location of predictions during evaluation, we also computed results at the sextant-level, where the prostate was divided into sextants according to standard systematic 12-core biopsy procedure. At the sextant-level, EPENet achieved mean sensitivity 61.1% and mean specificity 58.3%. Our approach has the potential to provide the location of extraprostatic extension using MRI alone, thus serving as an independent diagnostic aid to radiologists and facilitating treatment planning.

Bridging the gap between prostate radiology and pathology through machine learning.

BACKGROUND: Prostate cancer remains the second deadliest cancer for American men despite clinical advancements. Currently, magnetic resonance imaging (MRI) is considered the most sensitive non-invasive imaging modality that enables visualization, detection, and localization of prostate cancer, and is increasingly used to guide targeted biopsies for prostate cancer diagnosis. However, its utility remains limited due to high rates of false positives and false negatives as well as low inter-reader agreements. PURPOSE: Machine learning methods to detect and localize cancer on prostate MRI can help standardize radiologist interpretations. However, existing machine learning methods vary not only in model architecture, but also in the ground truth labeling strategies used for model training. We compare different labeling strategies and the effects they have on the performance of different machine learning models for prostate cancer detection on MRI. METHODS: Four different deep learning models (SPCNet, U-Net, branched U-Net, and DeepLabv3+) were trained to detect prostate cancer on MRI using 75 patients with radical prostatectomy, and evaluated using 40 patients with radical prostatectomy and 275 patients with targeted biopsy. Each deep learning model was trained with four different label types: pathology-confirmed radiologist labels, pathologist labels on whole-mount histopathology images, and lesion-level and pixel-level digital pathologist labels (previously validated deep learning algorithm on histopathology images to predict pixel-level Gleason patterns) on whole-mount histopathology images. The pathologist and digital pathologist labels (collectively referred to as pathology labels) were mapped onto pre-operative MRI using an automated MRI-histopathology registration platform. RESULTS: Radiologist labels missed cancers (ROC-AUC: 0.75-0.84), had lower lesion volumes (~68% of pathology lesions), and lower Dice overlaps (0.24-0.28) when compared with pathology labels. Consequently, machine learning models trained with radiologist labels also showed inferior performance compared to models trained with pathology labels. Digital pathologist labels showed high concordance with pathologist labels of cancer (lesion ROC-AUC: 0.97-1, lesion Dice: 0.75-0.93). Machine learning models trained with digital pathologist labels had the highest lesion detection rates in the radical prostatectomy cohort (aggressive lesion ROC-AUC: 0.91-0.94), and had generalizable and comparable performance to pathologist label-trained-models in the targeted biopsy cohort (aggressive lesion ROC-AUC: 0.87-0.88), irrespective of the deep learning architecture. Moreover, machine learning models trained with pixel-level digital pathologist labels were able to selectively identify aggressive and indolent cancer components in mixed lesions on MRI, which is not possible with any human-annotated label type. CONCLUSIONS: Machine learning models for prostate MRI interpretation that are trained with digital pathologist labels showed higher or comparable performance with pathologist label-trained models in both radical prostatectomy and targeted biopsy cohort. Digital pathologist labels can reduce challenges associated with human annotations, including labor, time, inter- and intra-reader variability, and can help bridge the gap between prostate radiology and pathology by enabling the training of reliable machine learning models to detect and localize prostate cancer on MRI.

Selective identification and localization of indolent and aggressive prostate cancers via CorrSigNIA: an MRI-pathology correlation and deep learning framework.

Automated methods for detecting prostate cancer and distinguishing indolent from aggressive disease on Magnetic Resonance Imaging (MRI) could assist in early diagnosis and treatment planning. Existing automated methods of prostate cancer detection mostly rely on ground truth labels with limited accuracy, ignore disease pathology characteristics observed on resected tissue, and cannot selectively identify aggressive (Gleason Pattern≥4) and indolent (Gleason Pattern=3) cancers when they co-exist in mixed lesions. In this paper, we present a radiology-pathology fusion approach, CorrSigNIA, for the selective identification and localization of indolent and aggressive prostate cancer on MRI. CorrSigNIA uses registered MRI and whole-mount histopathology images from radical prostatectomy patients to derive accurate ground truth labels and learn correlated features between radiology and pathology images. These correlated features are then used in a convolutional neural network architecture to detect and localize normal tissue, indolent cancer, and aggressive cancer on prostate MRI. CorrSigNIA was trained and validated on a dataset of 98 men, including 74 men that underwent radical prostatectomy and 24 men with normal prostate MRI. CorrSigNIA was tested on three independent test sets including 55 men that underwent radical prostatectomy, 275 men that underwent targeted biopsies, and 15 men with normal prostate MRI. CorrSigNIA achieved an accuracy of 80% in distinguishing between men with and without cancer, a lesion-level ROC-AUC of 0.81±0.31 in detecting cancers in both radical prostatectomy and biopsy cohort patients, and lesion-levels ROC-AUCs of 0.82±0.31 and 0.86±0.26 in detecting clinically significant cancers in radical prostatectomy and biopsy cohort patients respectively. CorrSigNIA consistently outperformed other methods across different evaluation metrics and cohorts. In clinical settings, CorrSigNIA may be used in prostate cancer detection as well as in selective identification of indolent and aggressive components of prostate cancer, thereby improving prostate cancer care by helping guide targeted biopsies, reducing unnecessary biopsies, and selecting and planning treatment.

Can patient decision aids reduce decisional conflict in a de-escalation of breast radiotherapy clinical trial? The PRIMETIME Study Within a Trial implemented using a cluster stepped-wedge trial design.

BACKGROUND: For patients with early breast cancer considered at very-low risk of local relapse, risks of radiotherapy may outweigh the benefits. Decisions regarding treatment omission can lead to patient uncertainty (decisional conflict), which may be lessened with patient decision aids (PDA). PRIMETIME (ISRCTN 41579286) is a UK-led biomarker-directed study evaluating omission of adjuvant radiotherapy in breast cancer; an embedded Study Within A Trial (SWAT) investigated whether PDA reduces decisional conflict using a cluster stepped-wedge trial design. METHODS: PDA diagrams and a video explaining risks and benefits of radiotherapy were developed in close collaboration between patient advocates and PRIMETIME trialists. The SWAT used a cluster stepped-wedge trial design, where each cluster represented the radiotherapy centre and referring peripheral centres. All clusters began in the standard information group (patient information and diagrams) and were randomised to cross-over to the enhanced information group (standard information plus video) at 2, 4 or 6 months. Primary endpoint was the decisional conflict scale (0-100, higher scores indicating greater conflict) which was assessed on an individual participant level. Multilevel mixed effects models used a random effect for cluster and a fixed effect for each step to adjust for calendar time and clustering. Robust standard errors were also adjusted for the clustering effect. RESULTS: Five hundred twenty-one evaluable questionnaires were returned from 809 eligible patients (64%) in 24 clusters between April 2018 and October 2019. Mean decisional conflict scores in the standard group (N = 184) were 10.88 (SD 11.82) and 8.99 (SD 11.82) in the enhanced group (N = 337), with no statistically significant difference [mean difference - 1.78, 95%CI - 3.82-0.25, p = 0.09]. Compliance with patient information and diagrams was high in both groups although in the enhanced group only 121/337 (36%) reported watching the video. CONCLUSION: The low levels of decisional conflict in PRIMETIME are reassuring and may reflect the high-quality information provision, such that not everyone required the video. This reinforces the importance of working with patients as partners in clinical trials especially in the development of patient-centred information and decision aids.

Deep Learning Improves Speed and Accuracy of Prostate Gland Segmentations on Magnetic Resonance Imaging for Targeted Biopsy.

PURPOSE: Targeted biopsy improves prostate cancer diagnosis. Accurate prostate segmentation on magnetic resonance imaging (MRI) is critical for accurate biopsy. Manual gland segmentation is tedious and time-consuming. We sought to develop a deep learning model to rapidly and accurately segment the prostate on MRI and to implement it as part of routine magnetic resonance-ultrasound fusion biopsy in the clinic. MATERIALS AND METHODS: A total of 905 subjects underwent multiparametric MRI at 29 institutions, followed by magnetic resonance-ultrasound fusion biopsy at 1 institution. A urologic oncology expert segmented the prostate on axial T2-weighted MRI scans. We trained a deep learning model, ProGNet, on 805 cases. We retrospectively tested ProGNet on 100 independent internal and 56 external cases. We prospectively implemented ProGNet as part of the fusion biopsy procedure for 11 patients. We compared ProGNet performance to 2 deep learning networks (U-Net and holistically-nested edge detector) and radiology technicians. The Dice similarity coefficient (DSC) was used to measure overlap with expert segmentations. DSCs were compared using paired t-tests. RESULTS: ProGNet (DSC=0.92) outperformed U-Net (DSC=0.85, p <0.0001), holistically-nested edge detector (DSC=0.80, p <0.0001), and radiology technicians (DSC=0.89, p <0.0001) in the retrospective internal test set. In the prospective cohort, ProGNet (DSC=0.93) outperformed radiology technicians (DSC=0.90, p <0.0001). ProGNet took just 35 seconds per case (vs 10 minutes for radiology technicians) to yield a clinically utilizable segmentation file. CONCLUSIONS: This is the first study to employ a deep learning model for prostate gland segmentation for targeted biopsy in routine urological clinical practice, while reporting results and releasing the code online. Prospective and retrospective evaluations revealed increased speed and accuracy.

Automated detection of aggressive and indolent prostate cancer on magnetic resonance imaging.

PURPOSE: While multi-parametric magnetic resonance imaging (MRI) shows great promise in assisting with prostate cancer diagnosis and localization, subtle differences in appearance between cancer and normal tissue lead to many false positive and false negative interpretations by radiologists. We sought to automatically detect aggressive cancer (Gleason pattern ≥ 4) and indolent cancer (Gleason pattern 3) on a per-pixel basis on MRI to facilitate the targeting of aggressive cancer during biopsy. METHODS: We created the Stanford Prostate Cancer Network (SPCNet), a convolutional neural network model, trained to distinguish between aggressive cancer, indolent cancer, and normal tissue on MRI. Ground truth cancer labels were obtained by registering MRI with whole-mount digital histopathology images from patients who underwent radical prostatectomy. Before registration, these histopathology images were automatically annotated to show Gleason patterns on a per-pixel basis. The model was trained on data from 78 patients who underwent radical prostatectomy and 24 patients without prostate cancer. The model was evaluated on a pixel and lesion level in 322 patients, including six patients with normal MRI and no cancer, 23 patients who underwent radical prostatectomy, and 293 patients who underwent biopsy. Moreover, we assessed the ability of our model to detect clinically significant cancer (lesions with an aggressive component) and compared it to the performance of radiologists. RESULTS: Our model detected clinically significant lesions with an area under the receiver operator characteristics curve of 0.75 for radical prostatectomy patients and 0.80 for biopsy patients. Moreover, the model detected up to 18% of lesions missed by radiologists, and overall had a sensitivity and specificity that approached that of radiologists in detecting clinically significant cancer. CONCLUSIONS: Our SPCNet model accurately detected aggressive prostate cancer. Its performance approached that of radiologists, and it helped identify lesions otherwise missed by radiologists. Our model has the potential to assist physicians in specifically targeting the aggressive component of prostate cancers during biopsy or focal treatment.

Classifying the emotional speech content of participants in group meetings using convolutional long short-term memory network.

Emotion is a central component of verbal communication between humans. Due to advances in machine learning and the development of affective computing, automatic emotion recognition is increasingly possible and sought after. To examine the connection between emotional speech and significant group dynamics perceptions, such as leadership and contribution, a new dataset (14 group meetings, 45 participants) is collected for analyzing collaborative group work based on the lunar survival task. To establish a training database, each participant's audio is manually annotated both categorically and along a three-dimensional scale with axes of activation, dominance, and valence and then converted to spectrograms. The performance of several neural network architectures for predicting speech emotion are compared for two tasks: categorical emotion classification and 3D emotion regression using multitask learning. Pretraining each neural network architecture on the well-known IEMOCAP (Interactive Emotional Dyadic Motion Capture) corpus improves the performance on this new group dynamics dataset. For both tasks, the two-dimensional convolutional long short-term memory network achieves the highest overall performance. By regressing the annotated emotions against post-task questionnaire variables for each participant, it is shown that the emotional speech content of a meeting can predict 71% of perceived group leaders and 86% of major contributors.

Direct matrix isolation IR spectroscopic evidence of halogen bonding from a comparative study of complexes of CBr4 and CCl4with acetone and formic acid.

Binary complexes of acetone and formic acid with tetrahalomethanes CBr4 and CCl4 have been isolated in argon matrix. Spectral shifts in the characteristic νC=O region of acetone, as well as in the fingerprint regions, are unambiguously assigned to the formation of halogen bond involving one of the halogen atoms on CBr4/CCl4 as donor, and the carbonyl oxygen of acetone as acceptor. The higher magnitude of shifts of νC=O and the fingerprint vibrations for the CBr4 complex, as compared to the CCl4 complex, is consistent with theoretical predictions of higher value of positive electrostatic potential in the "σ-hole" region of the former, and hence its higher susceptibility to halogen bonding. The formation of halogen bonded complexes involving formic acid as acceptor and CBr4/CCl4 as donors is also being reported for the first time. In this case too, distinct shifts are obtained in the νC=O as well as νC-O regions of formic acid, which again are significantly larger in magnitude for the CBr4 complex, as compared to the CCl4 complex. Electronic structure calculations have been carried out using different theoretical methods to identify the various possible structural isomers of the halogen bonded complexes, and to obtain relevant information regarding their energies and intermolecular geometrical parameters. In addition, NBO and AIM analysis have been carried out to understand the role of local interactions at the halogen bonded interface. Such predicted data are found to be consistent with experimental observations, and re-assert the stronger nature of CBr4 as halogen bond donor, as compared to CCl4.

Medium-Dependent Crossover from the Red to Blue Shift of the Donor's Stretching Fundamental in the Binary Hydrogen-Bonded Complexes of CDCl3 with Ethers and Ketones.

Mid-infrared spectra for C-D···O hydrogen (H)-bonded binary complexes of CDCl3 with acetone (AC), cyclohexanone (CHN), diethyl ether (DEE), and tetrahydrofuran (THF) have been measured in the vapor phase at room temperature and in an argon matrix at 8 K. Remarkable matrix effect has been observed in each case with respect to the spectral shift of the donor group's stretching fundamental (ΔνC-D). In the case of complexes with AC and CHN, the sign of ΔνC-D changes from a few wavenumbers positive (blue shift) in the vapor phase to a few tens of wavenumbers negative (red shift) in the argon matrix. For the two ether complexes, although no apparent reversal in the sign of ΔνC-D occurs, but the magnitudes of the red shifts in the matrix are manifold larger, and the bands appear with large enhancement in transition intensity. The medium effect has been explained consistently in terms of the local hyperconjugative charge transfer interaction at the H-bonding sites of the complexes and its interplay with the H-bond distance that varies with the physical conditions of the medium. Under the matrix isolation condition, νC-D bands of CHN and THF complexes depict a large number of substructures, which has been interpreted in terms of matrix site effect as well as Fermi resonance enhancement of the fingerprint combination tones and trapping of more than one isomer of the complexes in the matrix sites.

Identification of an Emitting Metastable State of p-Fluorophenol-Ammonia 1:2 Complex by Laser-Induced Fluorescence Spectroscopy.

We have demonstrated here, for the first time to our knowledge, the formation of an emitting metastable species upon lowest electronic excitation (S1) of a hydrogen-bonded 1:2 complex of para-fluorophenol (pFP) with ammonia (NH3), which is known to be one of the smallest reactive complexes to undergo excited state H-atom transfer (HAT) reaction to produce •NH4(NH3) radical fragment. The emission spectrum of the species is characterized to be red-shifted, broad, and structureless. From the viewpoint of energy balance, an excited state proton transfer (ESPT) is unfavorable, but according to predicted electronic structure parameters, the metastable state species could be stabilized by charge transfer (CT) interaction at the hydrogen-bonded geometry of the complex. We propose that this species could act as an intermediate to the HAT process in the excited state. The observation of such a state could be valuable to understand the complex dynamics of similar events in biologically relevant systems.

Hypoxia and angiogenic biomarkers in prostate cancer after external beam radiotherapy (EBRT) alone or combined with high-dose-rate brachytherapy boost (HDR-BTb).

PURPOSE: To investigate angiogenic and hypoxia biomarkers to predict outcome in patients receiving external beam radiotherapy (EBRT) alone or combined with high-dose-rate brachytherapy boost (HDR-BTb) for localised prostate cancer. METHODS: Prostate biopsy samples were collected prospectively in patients entered into a phase 3 randomised controlled trial of patients receiving EBRT or EBRT + HDR-BTb. Univariate and multivariate analyses using Cox proportional hazards model were performed to identify associations between immunohistochemical staining of hypoxia inducible factor 1 alpha (HIF1α), glucose transporter 1 (GLUT1), osteopontin (OPN) and microvessel density (MVD) using CD-34 antibody with clinical outcome. The primary endpoint was biochemical relapse free survival (BRFS) and secondary endpoint was distant metastasis free survival (DMFS). RESULTS: Immunohistochemistry was available for 204 patients. Increased OPN (Hazard ratio [HR] 2.38, 95% Confidence Interval [CI] 1.06-5.34, p < 0.036) and GLUT1 (HR 2.36, 95%CI 1.39-4.01, p < 0.001) expression were predictive of worse BRFS. Increased GLUT1 expression (HR 2.22, 1.02-4.84, p = 0.045) was predictive of worse DMFS. Increased MVD (CD-34) (HR 1.82, 95%CI 1.06-3.14, p = 0.03) and OPN (HR 1.82, 95%CI 1.06-3.14, p = 0.03) but reduced GLUT1 expression (HR 0.40, 95%CI 0.20-0.79, p = 0.009) were predictive of improved BRFS in patients receiving EBRT + HDR-BTb. CONCLUSION: Our data suggest angiogenic and hypoxia biomarkers may predict outcome and benefit of dose escalation, however further validation in prospective studies including hypoxia modification is needed. Trial registration number ISRCTN98241100, registered with ISRCTN at http://www.controlled-trials.com/isrctn/.