Discovering and interpreting transcriptomic drivers of imaging traits using neural networks

Evaluating an information theoretic approach for selecting multimodal data fusion methods

OBJECTIVE

Interest has grown in combining radiology, pathology, genomic, and clinical data to improve the accuracy of diagnostic and prognostic predictions toward precision health. However, most existing works choose their datasets and modeling approaches empirically and in an ad hoc manner. A prior study proposed four partial information decomposition (PID)-based metrics to provide a theoretical understanding of multimodal data interactions: redundancy, uniqueness of each modality, and synergy. However, these metrics have only been evaluated in a limited collection of biomedical data, and the existing work does not elucidate the effect of parameter selection when calculating the PID metrics. In this work, we evaluate PID metrics on a wider range of biomedical data, including clinical, radiology, pathology, and genomic data, and propose potential improvements to the PID metrics.

METHODS

We apply the PID metrics to seven different modality pairs across four distinct cohorts (datasets). We compare and interpret trends in the resulting PID metrics and downstream model performance in these multimodal cohorts. The downstream tasks being evaluated include predicting the prognosis (either overall survival or recurrence) of patients with non-small cell lung cancer, prostate cancer, and glioblastoma.

RESULTS

We found that, while PID metrics are informative, solely relying on these metrics to decide on a fusion approach does not always yield a machine learning model with optimal performance. Of the seven different modality pairs, three had poor (0%), three had moderate (66%-89%), and only one had perfect (100%) consistency between the PID values and model performance. We propose two improvements to the PID metrics (determining the optimal parameters and uncertainty estimation) and identified areas where PID metrics could be further improved.

CONCLUSION

The current PID metrics are not accurate enough for estimating the multimodal data interactions and need to be improved before they can serve as a reliable tool. We propose improvements and provide suggestions for future work. Code: https://github.com/zhtyolivia/pid-multimodal.

Cross-attention enables deep learning on limited omics-imaging-clinical data of 130 lung cancer patients

Deep-learning tools that extract prognostic factors derived from multi-omics data have recently contributed to individualized predictions of survival outcomes. However, the limited size of integrated omics-imaging-clinical datasets poses challenges. Here, we propose two biologically interpretable and robust deep-learning architectures for survival prediction of non-small cell lung cancer (NSCLC) patients, learning simultaneously from computed tomography (CT) scan images, gene expression data, and clinical information. The proposed models integrate patient-specific clinical, transcriptomic, and imaging data and incorporate Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome pathway information, adding biological knowledge within the learning process to extract prognostic gene biomarkers and molecular pathways. While both models accurately stratify patients in high- and low-risk groups when trained on a dataset of only 130 patients, introducing a cross-attention mechanism in a sparse autoencoder significantly improves the performance, highlighting tumor regions and NSCLC-related genes as potential biomarkers and thus offering a significant methodological advancement when learning from small imaging-omics-clinical samples.

Radiogenomic Predictors of Recurrence in Glioblastoma—A Systematic Review

Glioblastoma, as the most aggressive brain tumor, is associated with a poor prognosis and outcome. To optimize prognosis and clinical therapy decisions, there is an urgent need to stratify patients with increased risk for recurrent tumors and low therapeutic success to optimize individual treatment. Radiogenomics establishes a link between radiological and pathological information. This review provides a state-of-the-art picture illustrating the latest developments in the use of radiogenomic markers regarding prognosis and their potential for monitoring recurrence. Databases PubMed, Google Scholar, and Cochrane Library were searched. Inclusion criteria were defined as diagnosis of glioblastoma with histopathological and radiological follow-up. Out of 321 reviewed articles, 43 articles met these inclusion criteria. Included studies were analyzed for the frequency of radiological and molecular tumor markers whereby radiogenomic associations were analyzed. Six main associations were described: radiogenomic prognosis, MGMT status, IDH, EGFR status, molecular subgroups, and tumor location. Prospective studies analyzing prognostic features of glioblastoma together with radiological features are lacking. By reviewing the progress in the development of radiogenomic markers, we provide insights into the potential efficacy of such an approach for clinical routine use eventually enabling early identification of glioblastoma recurrence and therefore supporting a further personalized monitoring and treatment strategy.

Extendable and explainable deep learning for pan-cancer radiogenomics research

Radiogenomics is a field where medical images and genomic profiles are jointly analyzed to answer critical clinical questions. Specifically, people want to identify non-invasive imaging biomarkers that are associated with both genomic features and clinical outcomes. Deep learning is an advanced computer science technique that has been applied in many fields, including medical image and genomic data analysis. This review summarizes the current state of deep learning in pan-cancer radiogenomic research, discusses its limitations, and indicates the potential future directions. Traditional machine learning in radiomics, genomics, and radiogenomics have also been briefly discussed. We also summarize the main pan-cancer radiogenomic research resources. Two characteristics of deep learning are emphasized when discussing its application to pan-cancer radiogenomics, which are extendibility and explainability.

Predictions, Pivots, and a Pandemic: a Review of 2020's Top Translational Bioinformatics Publications

OBJECTIVES

Provide an overview of the emerging themes and notable papers which were published in 2020 in the field of Bioinformatics and Translational Informatics (BTI) for the International Medical Informatics Association Yearbook.

METHODS

A team of 16 individuals scanned the literature from the past year. Using a scoring rubric, papers were evaluated on their novelty, importance, and objective quality. 1,224 Medical Subject Headings (MeSH) terms extracted from these papers were used to identify themes and research focuses. The authors then used the scoring results to select notable papers and trends presented in this manuscript.

RESULTS

The search phase identified 263 potential papers and central themes of coronavirus disease 2019 (COVID-19), machine learning, and bioinformatics were examined in greater detail.

CONCLUSIONS

When addressing a once in a centruy pandemic, scientists worldwide answered the call, with informaticians playing a critical role. Productivity and innovations reached new heights in both TBI and science, but significant research gaps remain.

Using deep neural networks and interpretability methods to identify gene expression patterns that predict radiomic features and histology in non-small cell lung cancer

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.