Melanoma is a lethal skin disease that develops from moles. This study aimed to integrate multimodal data to predict metastatic melanoma, which is highly aggressive and difficult to treat. The proposed EnsembleSKCM method evaluated the prediction performances of long noncoding RNAs (lncRNAs), protein-coding messenger genes (mRNAs) and pathology images (images) for metastatic melanoma. Feature selection was used to screen for metastatic biomarkers in the lncRNA and mRNA datasets. The integrated EnsembleSKCM model was built based on the weighted results of the lncRNA-, mRNA- and image-based models. EnsembleSKCM achieved 0.9444 in the prediction accuracy of metastatic melanoma and outperformed the single-modal prediction models based on the lncRNA, mRNA and image data. The experimental data suggest the importance of integrating the complementary information from the three data modalities. WGCNA was used to analyze the relationship of molecular-level features and image features, and the results show connections between them. Another cohort was used to validate our prediction.
Today, neural network models are widely used to predict whether a person will develop diabetes in the future. However, for fuzzy inference engine and Adaptive Network-based Fuzzy Inference System (ANFIS), it costs a lot when the number of features is large, and the model will be more explainable if we use feature selection methods. In this paper, we modify the ANFIS model, combine ANFIS and neural network, and propose the ANFIS-NN model. We use SMOTE to address the imbalance between different classes, and use RFE and Casual Inference to do feature selection work. Then, we train an ANFIS model, and use a 5-layer neural network to replace the last layers to improve prediction accuracy. Data comparison experiments shows that our models get 0.9812 on accuracy, 0.9790 on G-mean, 0.9776 on F1 score on PIMA dataset using Casual Inference feature selection method, much better than traditional ANFIS; and our model works better on other diabetes datasets. The ANFIS-NN proposed in this paper can also be applied to other datasets to predict diabetes.