Multi-Modal Diagnosis of Infectious Diseases in the Developing World

Abstract

In low and middle income countries, infectious diseases continue to have a significant impact, particularly amongst the poorest in society. Tetanus and hand foot and mouth disease (HFMD) are two such diseases and, in both, death is associated with autonomic nervous system dysfunction (ANSD). Currently, photoplethysmogram or electrocardiogram monitoring is used to detect deterioration in these patients, however expensive clinical monitors are often required. In this study, we employ low-cost and mobile wearable devices to collect patient vital signs unobtrusively; and we develop machine learning algorithms for automatic and rapid triage of patients that provide efficient use of clinical resources. Existing methods are mainly dependent on the prior detection of clinical features with limited exploitation of multimodal physiological data. Moreover, the latest developments in deep learning (e.g. cross-domain transfer learning) have not been sufficiently applied for infectious disease diagnosis. In this paper, we present a fusion of multi-modal physiological data to predict the severity of ANSD with a hierarchy of resource-aware decision making. First, an on-site triage process is performed using a simple classifier. Second, personalised longitudinal modelling is employed that takes the previous states of the patient into consideration. We have also employed a spectrogram representation of the physiological waveforms to exploit existing networks for cross-domain transfer learning, which avoids the laborious and data intensive process of training a network from scratch. Results show that the proposed framework has promising potential in supporting severity grading of infectious diseases in low-resources settings, such as in the developing world.

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

“…Among the 25 studies applying feature-representation transfer, support vector machines and 'vanilla' neural networks were the most common methods used as the final model producing outputs based on the feature representations. Seven studies applied and compared multiple methods for this purpose [18][19][20][21][22][23][24].…”

“…E.g. models developed in the ImageNet dataset (>1 million images) were used even in smaller clinical studies with <100 patients [24,32,33], where the use of machine learning models would otherwise not be recommended or feasible. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review)…”

The Promise of Clinical Decision Support Systems Targetting Low-Resource Settings

“…Finally, the decision support is quite primitive and manifests itself in the form of reminders to order certain medication tests. Most recently, Tadesse et al [85] proposed a multi-modal approach to diagnose tetanus and hand foot and mouth disease (HFMD). They first convert electrocardiogram and photoplethysmogram time-series data to their spectrogram counterparts.…”

Identification of Neurodegenerative Diseases Based on Vertical Ground Reaction Force Classification Using Time–Frequency Spectrogram and Deep Learning Neural Network Features

“…They developed a dual-channel LSTM model to combine gait time series and force series recorded from NDD patients in order to understand the whole gait. According to several electrocardiogram (ECG) classification studies [ 24 , 25 , 26 , 27 ], combining a time–frequency representation (spectrogram) with a deep neural network can improve performance in extracting the distribution of important features more easily and automatically learn complex representation features directly from data. The goal of feature enrichment in a spectrogram of a time series signal is to enrich the simplified representation by restoring topological information, neighborhood information, and association information with details [ 28 ].…”