Table 5.

Contributions of recent studies integrating multi-modal/multi-omics data.

Author and Citation	Contributions
Schulz et al. [50]	Developed a multimodal deep learning model for prognosis prediction in clear-cell renal cell carcinoma.
Chen et al. [51]	Proposed pathomic fusion, a strategy for fusing histopathology and genomic features for improved cancer diagnosis and prognosis.
Zhang et al. [52]	Proposed a deep tensor survival model integrating multi-omics cancer data to improve cancer survival outcome prediction.
Malik et al. [53]	Integrated multi-omics data using a neural network framework to predict survival and drug response in breast cancer patients.
Hassanzadeh et al. [54]	Presented an integrated deep belief network that analyzes RNA, miRNA, and methylation molecular data to predict cancer survival and provide risk stratification.
Zhang et al. [55]	Presented OmiEmbed, a multi-task deep learning framework for multi-omics data.
Wei et al. [56]	Proposed a deep learning-based approach leveraging multi-omics data for biochemical relapse prediction in prostate dancer patients.
Karabacak et al. [57]	Utilized a CNN-based deep learning model to stratify low-grade gliomas using a multiple-gene signature and MRI data.
Park et al. [58]	Constructed a multi-omics data-affinitive artificial intelligence algorithm based on the graph convolutional network to predict non-small-cell lung cancer.
Steyaert et al. [59]	Developed a deep learning framework for multimodal data fusion for prognosis prediction in brain tumors.
Chen et al. [60]	Integrated radiomic features with genomic data to improve the survival analysis for non-small cell lung cancer patients.
Choi and Lee [61]	Developed Multi-PEN, a deep learning model for prognosis estimation in low-grade glioma patients.
Zhou et al. [62]	Developed a deep learning model to classify Nottingham prognostic index score levels for breast cancer patients, leveraging multi-omics data.
Islam et al. [63]	Proposed a radiogenomic overall survival prediction approach for GBM, integrating gene expression data with radiomic features.
Schmelz et al. [64]	Conducted in-depth analyses combining transcriptomic and genomic profiling in neuroblastoma patients, reporting continuous clonal evolution.
Yang et al. [65]	Developed HISMD, an immune subtyping system for HNSCC using multi-omics data and deep learning techniques on whole slide images.
Hira et al. [66]	Developed multi-omics analysis model for ovarian cancer using variational autoencoders.
Calabrese et al. [67]	Evaluated an artificial intelligence method for predicting clinically relevant genetic biomarkers from preoperative MRI in patients with glioblastoma.
Pan et al. [68]	Developed i-Modern, an integrated multi-omics deep learning network method, to identify potential therapeutic targets in glioma.
Tan et al. [69]	Presented a multi-modal fusion framework (MultiCoFusion) based on multi-task correlation learning for survival analysis and cancer grade classification.
Zhang et al. [70]	Conducted multi-omics data analyses to predict the prognosis of serous ovarian cancer (SOC) patients with principal component transformation (PCT).
Sharma et al. [71]	Developed a deep learning-based integrative model for survival time prediction in patients with HNSCC.
Tang et al. [72]	Developed a wavelet-based deep learning model for prognosis formulation in pancreatic adenocarcinoma.
Leng et al. [73]	Benchmarked deep learning methodologies for fusing multi-omics data, suggesting moGAT as the best performer for classification tasks, and efmmdVAE, efVAE, and IfmmdVAE for clustering tasks.
Carmichael et al. [74]	Proposed an integrative, exploratory analysis framework that uses angle-based joint.
Huang et al. [75]	Developed a model based on bidirectional deep neural networks (BiDNNs) to integrate DNA methylation and mRNA expression data for HCC samples.
Rescigno et al. [103]	Focused on characterizing CDK12-mutated mCRPC using a combination of targeted next-generation and exome sequencing techniques and deep learning.
Hu et al. [76]	Proposed a deep neural network, GCS-Net, for predicting gastric cancer prognosis based on biological information pathways.