Artificial intelligence, machine learning, and deep learning for clinical outcome prediction

Abstract

AI is a broad concept, grouping initiatives that use a computer to perform tasks that would usually require a human to complete. AI methods are well suited to predict clinical outcomes. In practice, AI methods can be thought of as functions that learn the outcomes accompanying standardized input data to produce accurate outcome predictions when trialed with new data. Current methods for cleaning, creating, accessing, extracting, augmenting, and representing data for training AI clinical prediction models are well defined. The use of AI to predict clinical outcomes is a dynamic and rapidly evolving arena, with new methods and applications emerging. Extraction or accession of electronic health care records and combining these with patient genetic data is an area of present attention, with tremendous potential for future growth. Machine learning approaches, including decision tree methods of Random Forest and XGBoost, and deep learning techniques including deep multi-layer and recurrent neural networks, afford unique capabilities to accurately create predictions from high dimensional, multimodal data. Furthermore, AI methods are increasing our ability to accurately predict clinical outcomes that previously were difficult to model, including time-dependent and multi-class outcomes. Barriers to robust AI-based clinical outcome model deployment include changing AI product development interfaces, the specificity of regulation requirements, and limitations in ensuring model interpretability, generalizability, and adaptability over time.

Machine-learning algorithm to non-invasively detect diabetes and pre-diabetes from electrocardiogram

“…This algorithm has several advantages like hardware optimisation and speed, a parallelised tree-building process, tree pruning using a depth-first approach and, most importantly, L1 and L2 regularisation to reduce or avert overfitting. The value of this increasingly popular algorithm in clinical classification tasks is demonstrated in two recent, elegant reviews 58 59. An important observation from our analyses was the performance superiority of XGBoost over mainstream DL models (online supplemental table 1).…”

The utility of machine learning for predicting donor discard in abdominal transplantation

“…We only tested one fully connected feedforward neural network architecture, which does not represent the full array of various architectures, regularization, techniques, and optimizers which could be utilized. However, given machine learning historic performance of equivalency or superiority on tabular datasets 59 such as the one that we have, as well as the significant improvement observed initially by machine learning models in our results, we felt comfortable moving forward with developing the random forest and XGBoost models and not pursuing further neural network architectures. Finally, decision tree-based models are generally more interpretable than deep learning models, as they maintain input feature representations, 59 which would add value for this clinically oriented classification task.…”

Predictors of shorter- and longer-term mortality after COVID-19 presentation among dialysis patients: parallel use of machine learning models in Latin and North American countries

“…Non-linear modeling, such as the machine learning techniques we utilized, can consider the effects for continuous variables without categorization and do not require arbitrary assumptions in linear relationships [25]. It is worthwhile to mention there have been advancements in predictive modeling techniques in recent years, and deep learning methods might have the potential to perform even better than the machine learning methods chosen by us due to the XGBoost model's ability to account for collinearity and missingness [28,29]. A limitation of these machine and deep learning models are that the outputs can be less intuitive on a population level.…”