Radiogenomic studies have suggested that biological heterogeneity of tumors is reflected radiographically through visible features on magnetic resonance (MR) images. We apply deep learning techniques to map between tumor gene expression profiles and tumor morphology in pre-operative MR studies of glioblastoma patients. A deep autoencoder was trained on 528 patients, each with 12,042 gene expressions. Then, the autoencoder's weights were used to initialize a supervised deep neural network. The supervised model was trained using a subset of 109 patients with both gene and MR data. For each patient, 20 morphological image features were extracted from contrast-enhancing and peritumoral edema regions. We found that neural network pre-trained with an autoencoder and dropout had lower errors than linear regression in predicting tumor morphology features by an average of 16.98% mean absolute percent error and 0.0114 mean absolute error, where several features were significantly different (adjusted p-value < 0.05). These results indicate neural networks, which can incorporate nonlinear, hierarchical relationships between gene expressions, may have the representational power to find more predictive radiogenomic associations than pairwise or linear methods.

Purpose: Integrative analysis combining diagnostic imaging and genomic information can uncover biological insights into lesions that are visible on radiologic images. We investigate techniques for interrogating a deep neural network trained to predict quantitative image (radiomic) features and histology from gene expression in non-small cell lung cancer (NSCLC). Approach: Using 262 training and 89 testing cases from two public datasets, deep feedforward neural networks were trained to predict the values of 101 computed tomography (CT) radiomic features and histology. A model interrogation method called gene masking was used to derive the learned associations between subsets of genes and a radiomic feature or histology class [adenocarcinoma (ADC), squamous cell, and other]. Results: Overall, neural networks outperformed other classifiers. In testing, neural networks classified histology with area under the receiver operating characteristic curves (AUCs) of 0.86 (ADC), 0.91 (squamous cell), and 0.71 (other). Classification performance of radiomics features ranged from 0.42 to 0.89 AUC. Gene masking analysis revealed new and previously reported associations. For example, hypoxia genes predicted histology ( >0.90  AUC ). Previously published gene signatures for classifying histology were also predictive in our model ( >0.80  AUC ). Gene sets related to the immune or cardiac systems and cell development processes were predictive ( >0.70  AUC ) of several different radiomic features. AKT signaling, tumor necrosis factor, and Rho gene sets were each predictive of tumor textures. Conclusions: This work demonstrates neural networks' ability to map gene expressions to radiomic features and histology types in NSCLC and to interpret the models to identify predictive genes associated with each feature or type.

Motivation

Cancer heterogeneity is observed at multiple biological levels. To improve our understanding of these differences and their relevance in medicine, approaches to link organ- and tissue-level information from diagnostic images and cellular-level information from genomics are needed. However, these 'radiogenomic' studies often use linear or shallow models, depend on feature selection, or consider one gene at a time to map images to genes. Moreover, no study has systematically attempted to understand the molecular basis of imaging traits based on the interpretation of what the neural network has learned. These studies are thus limited in their ability to understand the transcriptomic drivers of imaging traits, which could provide additional context for determining clinical outcomes.

Results

We present a neural network-based approach that takes high-dimensional gene expression data as input and performs non-linear mapping to an imaging trait. To interpret the models, we propose gene masking and gene saliency to extract learned relationships from radiogenomic neural networks. In glioblastoma patients, our models outperformed comparable classifiers (>0.10 AUC) and our interpretation methods were validated using a similar model to identify known relationships between genes and molecular subtypes. We found that tumor imaging traits had specific transcription patterns, e.g. edema and genes related to cellular invasion, and 10 radiogenomic traits were significantly predictive of survival. We demonstrate that neural networks can model transcriptomic heterogeneity to reflect differences in imaging and can be used to derive radiogenomic traits with clinical value.

Availability and implementation

https://github.com/novasmedley/deepRadiogenomics.

Contact

[email protected].

Supplementary information

Supplementary data are available at Bioinformatics online.

The growing amount of longitudinal data for a large population of patients has necessitated the application of algorithms that can discover patterns to inform patient management. This study demonstrates how temporal patterns generated from a combination of clinical and imaging measurements improve residual survival prediction in glioblastoma patients. Temporal patterns were identified with sequential pattern mining using data from 304 patients. Along with patient covariates, the patterns were incorporated as features in logistic regression models to predict 2-, 6-, or 9-month residual survival at each visit. The modeling approach that included temporal patterns achieved test performances of 0.820, 0.785, and 0.783 area under the receiver operating characteristic curve for predicting 2-, 6-, and 9-month residual survival, respectively. This approach significantly outperformed models that used tumor volume alone (p < 0.001) or tumor volume combined with patient covariates (p < 0.001) in training. Temporal patterns involving an increase in tumor volume above 122 mm³/day, a decrease in KPS across multiple visits, moderate neurologic symptoms, and worsening overall neurologic function suggested lower residual survival. These patterns are readily interpretable and found to be consistent with known prognostic indicators, suggesting they can provide early indicators to clinicians of changes in patient state and inform management decisions.

OBJECTIVE

It is crucial for clinicians to stay up to date on current literature in order to apply recent evidence to clinical decision making. Automatic summarization systems can help clinicians quickly view an aggregated summary of literature on a topic. Casama, a representation and summarization system based on “contextualized semantic maps,” captures the findings of biomedical studies as well as the contexts associated with patient population and study design. This paper presents a user-oriented evaluation of Casama in comparison to a context-free representation, SemRep.

MATERIALS AND METHODS

The effectiveness of the representation was evaluated by presenting users with manually annotated Casama and SemRep summaries of ten articles on driver mutations in cancer. Automatic annotations were evaluated on a collection of articles on EGFR mutation in lung cancer. Seven users completed a questionnaire rating the summarization quality for various topics and applications.

RESULTS

Casama had higher median scores than SemRep for the majority of the topics (p ≤ 0.00032), all of the applications (p ≤ 0.00089), and in overall summarization quality (p ≤ 1.5e-05). Casama’s manual annotations outperformed Casama’s automatic annotations (p = 0.00061).

DISCUSSION

Casama performed particularly well in the representation of strength of evidence, which was highly rated both quantitatively and qualitatively. Users noted that Casama’s less granular, more targeted representation improved usability compared to SemRep.

CONCLUSION

This evaluation demonstrated the benefits of a contextualized representation for summarizing biomedical literature on cancer. Iteration on specific areas of Casama’s representation, further development of its algorithms, and a clinically-oriented evaluation are warranted.

A search for proton decay into three charged leptons has been performed by using 0.37 Mton·years of data collected in Super-Kamiokande. All possible combinations of electrons, muons, and their antiparticles consistent with charge conservation were considered as decay modes. No significant excess of events has been found over the background, and lower limits on the proton lifetime divided by the branching ratio have been obtained. The limits range between 9.2×1033 and 3.4×1034 years at 90% confidence level, improving by more than an order of magnitude upon limits from previous experiments. A first limit has been set for the p→μ-e+e+ mode.

We have searched for proton decay via p→e+π0 and p→μ+π0 modes with the enlarged fiducial volume data of Super-Kamiokande from April 1996 to May 2018, which corresponds to 450 kton·years exposure. We have accumulated about 25% more livetime and enlarged the fiducial volume of the Super-Kamiokande detector from 22.5 kton to 27.2 kton for this analysis, so that 144 kton·years of data, including 78 kton·years of additional fiducial volume data, has been newly analyzed. No candidates have been found for p→e+π0 and one candidate remains for p→μ+π0 in the conventional 22.5 kton fiducial volume and it is consistent with the atmospheric neutrino background prediction. We set lower limits on the partial lifetime for each of these modes: τ/B(p→e+π0)>2.4×1034 years and τ/B(p→μ+π0)>1.6×1034 years at 90% confidence level.

As a baryon number violating process with ΔB=2, neutron-antineutron oscillation (n→n¯) provides a unique test of baryon number conservation. We have performed a search for n→n¯ oscillation with bound neutrons in Super-Kamiokande, with the full dataset from its first four run periods, representing an exposure of 0.37 Mton-years. The search used a multivariate analysis trained on simulated n→n¯ events and atmospheric neutrino backgrounds and resulted in 11 candidate events with an expected background of 9.3 events. In the absence of statistically significant excess, we derived a lower limit on n¯ appearance lifetime in O16 nuclei of 3.6×1032 years and on the neutron-antineutron oscillation time of τn→n¯>4.7×108 s at 90% C.L.

The US neutrino community gathered at the Workshop on the Intermediate Neutrino Program (WINP) at Brookhaven National Laboratory February 4-6, 2015 to explore opportunities in neutrino physics over the next five to ten years. Scientists from particle, astroparticle and nuclear physics participated in the workshop. The workshop examined promising opportunities for neutrino physics in the intermediate term, including possible new small to mid-scale experiments, US contributions to large experiments, upgrades to existing experiments, R&D plans and theory. The workshop was organized into two sets of parallel working group sessions, divided by physics topics and technology. Physics working groups covered topics on Sterile Neutrinos, Neutrino Mixing, Neutrino Interactions, Neutrino Properties and Astrophysical Neutrinos. Technology sessions were organized into Theory, Short-Baseline Accelerator Neutrinos, Reactor Neutrinos, Detector R&D and Source, Cyclotron and Meson Decay at Rest sessions.This report summarizes discussion and conclusions from the workshop.