Journal of Biomedical Informatics最新文献

Cancer is a disease that causes many deaths worldwide. The treatment of cancer is first and foremost a matter of detection, a treatment that is most effective when the disease is detected at an early stage. With the evolution of technology, several computer-aided diagnosis tools have been developed around cancer; several image-based cancer detection methods have been developed too. However, cancer detection faces many difficulties related to early detection which is crucial for patient survival rate. To detect cancer early, scientists have been using transcriptomic data. However, this presents some challenges such as unlabelled data, a large amount of data, and image-based techniques that only focus on one type of cancer. The purpose of this work is to develop a deep learning model that can effectively detect as soon as possible, specifically in the early stages, any type of cancer as an anomaly in transcriptomic data. This model must have the ability to act independently and not be restricted to any specific type of cancer. To achieve this goal, we modeled a deep neural network (a Variational Autoencoder) and then defined an algorithm for detecting anomalies in the output of the Variational Autoencoder. The Variational Autoencoder consists of an encoder and a decoder with a hidden layer. With the TCGA and GTEx data, we were able to train the model for six types of cancer using the Adam optimizer with decay learning for training, and a two-component loss function. As a result, we obtained the lowest value of accuracy 0.950, and the lowest value of recall 0.830. This research leads us to the design of a deep learning model for the detection of cancer as an anomaly in transcriptomic data.

Objective: Self-harm risk prediction models developed using health system data (electronic health records and insurance claims information) often use patient information from up to several years prior to the index visit when the prediction is made. Measurements from some time periods may not be available for all patients. Using the framework of algorithm-agnostic variable importance, we study the predictive potential of variables corresponding to different time horizons prior to the index visit and demonstrate the application of variable importance techniques in the biomedical informatics setting.

Materials and methods: We use variable importance to quantify the potential of recent (up to three months before the index visit) and distant (more than one year before the index visit) patient mental health information for predicting self-harm risk using data from seven health systems. We quantify importance as the decrease in predictiveness when the variable set of interest is excluded from the prediction task. We define predictiveness using discriminative metrics: area under the receiver operating characteristic curve (AUC), sensitivity, and positive predictive value.

Results: Mental health predictors corresponding to the three months prior to the index visit show strong signal of importance; in one setting, excluding these variables decreased AUC from 0.85 to 0.77. Predictors corresponding to more distant information were less important.

Discussion: Predictors from the months immediately preceding the index visit are highly important. Implementation of self-harm prediction models may be challenging in settings where recent data are not completely available (e.g., due to lags in insurance claims processing) at the time a prediction is made.

Conclusion: Clinically derived variables from different time frames exhibit varying levels of importance for predicting self-harm. Variable importance analyses can inform whether and how to implement risk prediction models into clinical practice given real-world data limitations. These analyses be applied more broadly in biomedical informatics research to provide insight into general clinical risk prediction tasks.