Deep Learning Predicts Hip Fracture using Confounding Patient and Healthcare Variables
Authors:
Marcus A. Badgeley,
John R. Zech,
Luke Oakden-Rayner,
Benjamin S. Glicksberg,
Manway Liu,
William Gale,
Michael V. McConnell,
Beth Percha,
Thomas M. Snyder,
Joel T. Dudley
Abstract:
Hip fractures are a leading cause of death and disability among older adults. Hip fractures are also the most commonly missed diagnosis on pelvic radiographs. Computer-Aided Diagnosis (CAD) algorithms have shown promise for hel** radiologists detect fractures, but the image features underpinning their predictions are notoriously difficult to understand. In this study, we trained deep learning mo…
▽ More
Hip fractures are a leading cause of death and disability among older adults. Hip fractures are also the most commonly missed diagnosis on pelvic radiographs. Computer-Aided Diagnosis (CAD) algorithms have shown promise for hel** radiologists detect fractures, but the image features underpinning their predictions are notoriously difficult to understand. In this study, we trained deep learning models on 17,587 radiographs to classify fracture, five patient traits, and 14 hospital process variables. All 20 variables could be predicted from a radiograph (p < 0.05), with the best performances on scanner model (AUC=1.00), scanner brand (AUC=0.98), and whether the order was marked "priority" (AUC=0.79). Fracture was predicted moderately well from the image (AUC=0.78) and better when combining image features with patient data (AUC=0.86, p=2e-9) or patient data plus hospital process features (AUC=0.91, p=1e-21). The model performance on a test set with matched patient variables was significantly lower than a random test set (AUC=0.67, p=0.003); and when the test set was matched on patient and image acquisition variables, the model performed randomly (AUC=0.52, 95% CI 0.46-0.58), indicating that these variables were the main source of the model's predictive ability overall. We also used Naive Bayes to combine evidence from image models with patient and hospital data and found their inclusion improved performance, but that this approach was nevertheless inferior to directly modeling all variables. If CAD algorithms are inexplicably leveraging patient and process variables in their predictions, it is unclear how radiologists should interpret their predictions in the context of other known patient data. Further research is needed to illuminate deep learning decision processes so that computers and clinicians can effectively cooperate.
△ Less
Submitted 8 November, 2018;
originally announced November 2018.
Transition from local to global phase synchrony in small world neural network and its possible implications for epilepsy
Authors:
Bethany Percha,
Rhonda Dzakpasu,
Michał Żochowski
Abstract:
Temporal correlations in the brain are thought to have very dichotomic roles. On one hand they are ubiquitously present in the healthy brain and are thought to underlie feature binding during information processing. On the other hand large scale synchronization is an underlying mechanism of epileptic seizures. In this paper we show a possible mechanism of transition to pathological coherence und…
▽ More
Temporal correlations in the brain are thought to have very dichotomic roles. On one hand they are ubiquitously present in the healthy brain and are thought to underlie feature binding during information processing. On the other hand large scale synchronization is an underlying mechanism of epileptic seizures. In this paper we show a possible mechanism of transition to pathological coherence underlying seizure generation. We show that properties of phase synchronization in the 2-D lattice of non-identical coupled Hindmarsh-Rose neurons change radically depending on the connectivity structure. We modify the connectivity using the small world network paradigm and measure properties of phase synchronization using previously developed measure based on assessment of the distributions of relative interspike intervals. We show that the phase synchronization undergoes a dramatic change as a function of locality of network connections from local coherence strongly dependent on the distance between two neurons to global coherence exhibiting stronger phase locking and spanning the whole network.
△ Less
Submitted 12 November, 2004;
originally announced November 2004.
