Combiner and HyperCombiner networks: Rules to combine multimodality MR images for prostate cancer localisation.

One of the distinct characteristics of radiologists reading multiparametric prostate MR scans, using reporting systems like PI-RADS v2.1, is to score individual types of MR modalities, including T2-weighted, diffusion-weighted, and dynamic contrast-enhanced, and then combine these image-modality-specific scores using standardised decision rules to predict the likelihood of clinically significant cancer. This work aims to demonstrate that it is feasible for low-dimensional parametric models to model such decision rules in the proposed Combiner networks, without compromising the accuracy of predicting radiologic labels. First, we demonstrate that either a linear mixture model or a nonlinear stacking model is sufficient to model PI-RADS decision rules for localising prostate cancer. Second, parameters of these combining models are proposed as hyperparameters, weighing independent representations of individual image modalities in the Combiner network training, as opposed to end-to-end modality ensemble. A HyperCombiner network is developed to train a single image segmentation network that can be conditioned on these hyperparameters during inference for much-improved efficiency. Experimental results based on 751 cases from 651 patients compare the proposed rule-modelling approaches with other commonly-adopted end-to-end networks, in this downstream application of automating radiologist labelling on multiparametric MR. By acquiring and interpreting the modality combining rules, specifically the linear-weights or odds ratios associated with individual image modalities, three clinical applications are quantitatively presented and contextualised in the prostate cancer segmentation application, including modality availability assessment, importance quantification and rule discovery.

Prototypical few-shot segmentation for cross-institution male pelvic structures with spatial registration.

The prowess that makes few-shot learning desirable in medical image analysis is the efficient use of the support image data, which are labelled to classify or segment new classes, a task that otherwise requires substantially more training images and expert annotations. This work describes a fully 3D prototypical few-shot segmentation algorithm, such that the trained networks can be effectively adapted to clinically interesting structures that are absent in training, using only a few labelled images from a different institute. First, to compensate for the widely recognised spatial variability between institutions in episodic adaptation of novel classes, a novel spatial registration mechanism is integrated into prototypical learning, consisting of a segmentation head and an spatial alignment module. Second, to assist the training with observed imperfect alignment, support mask conditioning module is proposed to further utilise the annotation available from the support images. Extensive experiments are presented in an application of segmenting eight anatomical structures important for interventional planning, using a data set of 589 pelvic T2-weighted MR images, acquired at seven institutes. The results demonstrate the efficacy in each of the 3D formulation, the spatial registration, and the support mask conditioning, all of which made positive contributions independently or collectively. Compared with the previously proposed 2D alternatives, the few-shot segmentation performance was improved with statistical significance, regardless whether the support data come from the same or different institutes.

Delineating COVID-19 subgroups using routine clinical data identifies distinct in-hospital outcomes.

The COVID-19 pandemic has been a great challenge to healthcare systems worldwide. It highlighted the need for robust predictive models which can be readily deployed to uncover heterogeneities in disease course, aid decision-making and prioritise treatment. We adapted an unsupervised data-driven model-SuStaIn, to be utilised for short-term infectious disease like COVID-19, based on 11 commonly recorded clinical measures. We used 1344 patients from the National COVID-19 Chest Imaging Database (NCCID), hospitalised for RT-PCR confirmed COVID-19 disease, splitting them equally into a training and an independent validation cohort. We discovered three COVID-19 subtypes (General Haemodynamic, Renal and Immunological) and introduced disease severity stages, both of which were predictive of distinct risks of in-hospital mortality or escalation of treatment, when analysed using Cox Proportional Hazards models. A low-risk Normal-appearing subtype was also discovered. The model and our full pipeline are available online and can be adapted for future outbreaks of COVID-19 or other infectious disease.

Prostate cancer segmentation from MRI by a multistream fusion encoder.

BACKGROUND: Targeted prostate biopsy guided by multiparametric magnetic resonance imaging (mpMRI) detects more clinically significant lesions than conventional systemic biopsy. Lesion segmentation is required for planning MRI-targeted biopsies. The requirement for integrating image features available in T2-weighted and diffusion-weighted images poses a challenge in prostate lesion segmentation from mpMRI. PURPOSE: A flexible and efficient multistream fusion encoder is proposed in this work to facilitate the multiscale fusion of features from multiple imaging streams. A patch-based loss function is introduced to improve the accuracy in segmenting small lesions. METHODS: The proposed multistream encoder fuses features extracted in the three imaging streams at each layer of the network, thereby allowing improved feature maps to propagate downstream and benefit segmentation performance. The fusion is achieved through a spatial attention map generated by optimally weighting the contribution of the convolution outputs from each stream. This design provides flexibility for the network to highlight image modalities according to their relative influence on the segmentation performance. The encoder also performs multiscale integration by highlighting the input feature maps (low-level features) with the spatial attention maps generated from convolution outputs (high-level features). The Dice similarity coefficient (DSC), serving as a cost function, is less sensitive to incorrect segmentation for small lesions. We address this issue by introducing a patch-based loss function that provides an average of the DSCs obtained from local image patches. This local average DSC is equally sensitive to large and small lesions, as the patch-based DSCs associated with small and large lesions have equal weights in this average DSC. RESULTS: The framework was evaluated in 931 sets of images acquired in several clinical studies at two centers in Hong Kong and the United Kingdom. In particular, the training, validation, and test sets contain 615, 144, and 172 sets of images, respectively. The proposed framework outperformed single-stream networks and three recently proposed multistream networks, attaining F1 scores of 82.2 and 87.6% in the lesion and patient levels, respectively. The average inference time for an axial image was 11.8 ms. CONCLUSION: The accuracy and efficiency afforded by the proposed framework would accelerate the MRI interpretation workflow of MRI-targeted biopsy and focal therapies.

Prediction of the Endocrine disruption profile of fluorinated biphenyls and analogues: An in silico study.

Fluorinated biphenyls and their analogues (FBAs) are considered new persistent organic pollutants, but their endocrine-disrupting effects are still unknown. To fill this gap, the binding probability of 44 FBAs to different nuclear hormone receptors (NHRs) was predicted using Endocrine Disruptome. And molecular similarity and network toxicology analysis were used to strengthen the docking screening. The docking results showed that FBAs could have high binding potential for various NHRs, such as estrogen receptors ß antagonism (ERß an), liver X receptors α (LXRα), estrogen receptors α (ERα), and liver X receptors ß (LXRß). The similarity analysis found that the degree of overlap of the NHR repertoire was related to the Tanimoto coefficient of FBAs. Network toxicology verified a part of docking screening results and identified endocrine-disrupting pathways worthy of attention. This study found out potential endocrine-disrupting FBAs and their vulnerable, and developed a workflow that would leverage in silico approaches including molecular docking, similarity, and network toxicology for risk prioritization of potential endocrine-disrupting compounds.

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.

Cross-Modality Image Registration Using a Training-Time Privileged Third Modality.

In this work, we consider the task of pairwise cross-modality image registration, which may benefit from exploiting additional images available only at training time from an additional modality that is different to those being registered. As an example, we focus on aligning intra-subject multiparametric Magnetic Resonance (mpMR) images, between T2-weighted (T2w) scans and diffusion-weighted scans with high b-value (DWI [Formula: see text]). For the application of localising tumours in mpMR images, diffusion scans with zero b-value (DWI [Formula: see text]) are considered easier to register to T2w due to the availability of corresponding features. We propose a learning from privileged modality algorithm, using a training-only imaging modality DWI [Formula: see text], to support the challenging multi-modality registration problems. We present experimental results based on 369 sets of 3D multiparametric MRI images from 356 prostate cancer patients and report, with statistical significance, a lowered median target registration error of 4.34 mm, when registering the holdout DWI [Formula: see text] and T2w image pairs, compared with that of 7.96 mm before registration. Results also show that the proposed learning-based registration networks enabled efficient registration with comparable or better accuracy, compared with a classical iterative algorithm and other tested learning-based methods with/without the additional modality. These compared algorithms also failed to produce any significantly improved alignment between DWI [Formula: see text] and T2w in this challenging application.

Task model-specific operator skill assessment in routine fetal ultrasound scanning.

PURPOSE: For highly operator-dependent ultrasound scanning, skill assessment approaches evaluate operator competence given available data, such as acquired images and tracked probe movement. Operator skill level can be quantified by the completeness, speed, and precision of performing a clinical task, such as biometry. Such clinical tasks are increasingly becoming assisted or even replaced by automated machine learning models. In addition to measurement, operators need to be competent at the upstream task of acquiring images of sufficient quality. To provide computer assistance for this task requires a new definition of skill. METHODS: This paper focuses on the task of selecting ultrasound frames for biometry, for which operator skill is assessed by quantifying how well the tasks are performed with neural network-based frame classifiers. We first develop a frame classification model for each biometry task, using a novel label-efficient training strategy. Once these task models are trained, we propose a second task model-specific network to predict two skill assessment scores, based on the probability of identifying positive frames and accuracy of model classification. RESULTS: We present comprehensive results to demonstrate the efficacy of both the frame-classification and skill-assessment networks, using clinically acquired data from two biometry tasks for a total of 139 subjects, and compare the proposed skill assessment with metrics of operator experience. CONCLUSION: Task model-specific skill assessment is feasible and can be predicted by the proposed neural networks, which provide objective assessment that is a stronger indicator of task model performance, compared to existing skill assessment methods.

Image quality assessment for machine learning tasks using meta-reinforcement learning.

In this paper, we consider image quality assessment (IQA) as a measure of how images are amenable with respect to a given downstream task, or task amenability. When the task is performed using machine learning algorithms, such as a neural-network-based task predictor for image classification or segmentation, the performance of the task predictor provides an objective estimate of task amenability. In this work, we use an IQA controller to predict the task amenability which, itself being parameterised by neural networks, can be trained simultaneously with the task predictor. We further develop a meta-reinforcement learning framework to improve the adaptability for both IQA controllers and task predictors, such that they can be fine-tuned efficiently on new datasets or meta-tasks. We demonstrate the efficacy of the proposed task-specific, adaptable IQA approach, using two clinical applications for ultrasound-guided prostate intervention and pneumonia detection on X-ray images.

Identification of a RAD52 Inhibitor Inducing Synthetic Lethality in BRCA2-Deficient Cancer Cells.

The breast cancer susceptibility gene 1/2 (BRCA1/2) is frequently mutated in many malignant tumors, such as breast cancer and ovarian cancer. Studies have demonstrated that inhibition of RAD52 gene function in BRCA2-deficient cancer causes synthetic lethality, suggesting a potential application of RAD52 in cancer-targeted therapy. In this study, we have performed a virtual screening by targeting the self-association domain (residues 85-159) of RAD52 with a library of 66,608 compounds and found one compound, C791-0064, that specifically inhibited the proliferation of BRCA2-deficient cancer cells. Our biochemical and cell-based experimental data suggested that C791-0064 specifically bound to RAD52 and disrupted the single-strand annealing activity of RAD52. Taken together, C791-0064 is a promising leading compound worthy of further exploitation in the context of BRCA-deficient targeted cancer therapy.

MRI Cross-Modality Image-to-Image Translation.

We present a cross-modality generation framework that learns to generate translated modalities from given modalities in MR images. Our proposed method performs Image Modality Translation (abbreviated as IMT) by means of a deep learning model that leverages conditional generative adversarial networks (cGANs). Our framework jointly exploits the low-level features (pixel-wise information) and high-level representations (e.g. brain tumors, brain structure like gray matter, etc.) between cross modalities which are important for resolving the challenging complexity in brain structures. Our framework can serve as an auxiliary method in medical use and has great application potential. Based on our proposed framework, we first propose a method for cross-modality registration by fusing the deformation fields to adopt the cross-modality information from translated modalities. Second, we propose an approach for MRI segmentation, translated multichannel segmentation (TMS), where given modalities, along with translated modalities, are segmented by fully convolutional networks (FCN) in a multichannel manner. Both of these two methods successfully adopt the cross-modality information to improve the performance without adding any extra data. Experiments demonstrate that our proposed framework advances the state-of-the-art on five brain MRI datasets. We also observe encouraging results in cross-modality registration and segmentation on some widely adopted brain datasets. Overall, our work can serve as an auxiliary method in medical use and be applied to various tasks in medical fields.

A novel EZH2 inhibitor induces synthetic lethality and apoptosis in PBRM1-deficient cancer cells.

The inhibition of enhancer of zeste homolog 2 (EZH2) has been suggested to be synthetic lethal with polybromo-1 (PBRM1) deficiency, rendering EZH2 to be an attractive target for the treatment of PBRM1 frequently mutated cancers. In the current study, we combined computational and biochemical approaches to establish an efficient system for the screening and validation of synthetic lethal inhibitors from a large pool of chemical compounds. Five putative EZH2 inhibitors were identified through structure-based virtual screening from 47,737 chemical compounds and analyzed with molecular dynamics. The efficacy of these compounds against EZH2 was tested using PBRM1 deficient and wide-type cell lines. The compound L501-1669 selectively inhibited the proliferation of PBRM1-deficient cells and down-regulated the tri-methylation of histone H3 at Lysine 27 (H3K27me3). Importantly, we also observed an increase in apoptotic activities in L501-1669 treated PBRM1-deficient cells. Taken together, our results demonstrate that L501-1669 is a selective EZH2 inhibitor with promising application in the targeted therapy of PBRM1-deficient cancers.

Aqueous extract of Fritillariae cirrhosae induces cellular apoptosis through activation of STATs-mediated immunomodulation.

ETHNOPHARMACOLOGICAL RELEVANCE: Fritillariae cirrhosae (FC), referred to'Chuan beimu'in China. As an important edible and medicinal plant, the bulbs of F.cirrhosae is used traditionally in the treatment of pulmonary diseases associated with lung heat, inflammation and tumors. In the study, we investigated the effect of aqueous extract of FC (FC-AE) and elucidated its mechanism in non-small cell lung cancer A549 cells and a xenograft model of nude mice. MATERIALS AND METHODS: CCK-8 and plate colony formation assay were used to evaluate the effect of FC-AE in A549 cells in vitro, and the gene expression profile of FC-AE on A549 cells was assessed by RNA sequencing system. Then, the effects of FC-AE on cell cycle and apoptosis of A549 cells were analyzed by flow cytometry. In combination with RNA-seq data, RT-PCR and western blot were used to evaluate the expression of proteins related to apoptosis and immune regulation. A xenograft model of nude mice was used to assess the effect of FC-AE in vivo. RESULTS: CCK-8 and plate cloning assays showed that FC-AE inhibited the proliferation and colony formation of A549 cells. A549 cells treated with FC-AE can triggered apoptosis. GO and KEGG pathway enrichment analysis of RNA-seq data showed that most of the differentially expressed genes (DEGs) were related to immune response, apoptosis and cell cycle process. Several immune and apoptotic DEGs were identified by qRT-PCR which were consistented with RNA-seq data. In nude mice, FC-AE reduced the tumor size and promoted the secretion of cytokines IL12 and IFNγ. FC-AE up-regulated the two members (STAT1 and STAT4) of STATs and their target genes (IFNγ and IL-12, respectively) protein expressions, and actively regulates Bcl-2/Bax family proteins which resulted in cellular apoptosis in A549 cells. CONCLUSION: Our finding suggests that FC-AE mediates apoptosis through a STAT1 and STAT4-mediated co-regulatory network, which may be the key novel mechanism for its antitumor activity. The F. cirrhosa may be a promising antitumor drug for modulating immune responses to improve cancer therapy.

Compound F779-0434 causes synthetic lethality in BRCA2-deficient cancer cells by disrupting RAD52-ssDNA association.

Maintenance of genomic integrity is essential for the survival of all organisms. Homologous recombination (HR) is the major pathway for high-fidelity repair of DNA double-stranded breaks (DSBs). In addition to the classic BRCA-RAD51 pathway, another secondary HR sub-pathway dependent on RAD52 has been suggested to be functioning in mammalian cells. Importantly, RAD52 has been shown to be synthetically lethal to BRCA1/2-deficient cells, rendering RAD52 to be a desirable target in cancer therapy. In the current study, we performed a structure-based virtual screening of 47 737 drug-like compounds to identify RAD52-specific inhibitors. The top ranked virtual screening hits were further characterized using molecular dynamics simulation and biochemical and cell-based assays. We found that one compound, namely, F779-0434 specifically suppressed the growth of BRCA2-deficient cells and disrupted RAD52-ssDNA interaction in vitro. This RAD52-specific inhibitor identified in the current study is a promising compound for targeted cancer therapy, and it can also be used as a probe to study the mechanisms of DNA repair in human cells.

L1198F Mutation Resensitizes Crizotinib to ALK by Altering the Conformation of Inhibitor and ATP Binding Sites.

The efficacy of anaplastic lymphoma kinase (ALK) positive non-small-cell lung cancer (NSCLC) treatment with small molecule inhibitors is greatly challenged by acquired resistance. A recent study reported the newest generation inhibitor resistant mutation L1198F led to the resensitization to crizotinib, which is the first Food and Drug Administration (FDA) approved drug for the treatment of ALK-positive NSCLC. It is of great importance to understand how this extremely rare event occurred for the purpose of overcoming the acquired resistance of such inhibitors. In this study, we exploited molecular dynamics (MD) simulation to dissect the molecular mechanisms. Our MD results revealed that L1198F mutation of ALK resulted in the conformational change at the inhibitor site and altered the binding affinity of ALK to crizotinib and lorlatinib. L1198F mutation also affected the autoactivation of ALK as supported by the identification of His1124 and Tyr1278 as critical amino acids involved in ATP binding and phosphorylation. Our findings are valuable for designing more specific and potent inhibitors for the treatment of ALK-positive NSCLC and other types of cancer.