Transfer learning for non-image data in clinical research: A scoping review

Abstract

Background Transfer learning is a form of machine learning where a pre-trained model trained on a specific task is reused as a starting point and tailored to another task in a different dataset. While transfer learning has garnered considerable attention in medical image analysis, its use for clinical non-image data is not well studied. Therefore, the objective of this scoping review was to explore the use of transfer learning for non-image data in the clinical literature. Methods and findings We systematically searched medical databases (PubMed, EMBASE, CINAHL) for peer-reviewed clinical studies that used transfer learning on human non-image data. We included 83 studies in the review. More than half of the studies (63%) were published within 12 months of the search. Transfer learning was most often applied to time series data (61%), followed by tabular data (18%), audio (12%) and text (8%). Thirty-three (40%) studies applied an image-based model to non-image data after transforming data into images (e.g. spectrograms). Twenty-nine (35%) studies did not have any authors with a health-related affiliation. Many studies used publicly available datasets (66%) and models (49%), but fewer shared their code (27%). Conclusions In this scoping review, we have described current trends in the use of transfer learning for non-image data in the clinical literature. We found that the use of transfer learning has grown rapidly within the last few years. We have identified studies and demonstrated the potential of transfer learning in clinical research in a wide range of medical specialties. More interdisciplinary collaborations and the wider adaption of reproducible research principles are needed to increase the impact of transfer learning in clinical research.

Prediction of Biliary Complications After Human Liver Transplantation Using Hyperspectral Imaging and Convolutional Neural Networks: A Proof-of-concept Study

“…To test the resulting CNN, we used the so-called leave-one-out cross-validation. 35,36 This technique is used for small data sets. It is based on the training of the network on n – 1 data sets (n = 14) and using the remaining data points as a test sample.…”

Automating the overburdened clinical coding system: challenges and next steps

“…Existing ACC models may potentially become inapplicable as ICD-11 and other updated code sets are implemented. Transitions of code sets could require new methods of data handling and mapping 14 , 15 .…”

Where do pigeons’ choices come from? Transfer Learning in Spiking Neural Networks for Animal Choices in Behavioral Research

“…However, transfer learning has yet to be well studied with non-image data in the clinical literature, where a limited number of studies have been published in a recent review (Ebbehoj et al, 2022), where the authors found that studies primarily used time-series data (n = 51, 61%), with ne-tuning being the most common transfer learning strategy (70%), followed by feature representation transfer (22%). Seven studies out of 83 used both approaches.…”