Deep-Precognitive Diagnosis: Preventing Future Pandemics by Novel Disease Detection With Biologically-Inspired
Conv-Fuzzy Network

Deep learning-based Computer-Aided Diagnosis has gained immense attention in recent years due to its capability to enhance diagnostic performance and elucidate complex clinical tasks. However, conventional supervised deep learning models are incapable of recognizing novel diseases that do not exist in the training dataset. Automated early-stage detection of novel infectious diseases can be vital in controlling their rapid spread. Moreover, the development of a conventional CAD model is only possible after disease outbreaks and datasets become available for training (viz. COVID-19 outbreak). Since novel diseases are unknown and cannot be included in training data, it is challenging to recognize them through existing supervised deep learning models. Even after data becomes available, recognizing new classes with conventional models requires a complete extensive re-training. The present study is the first to report this problem and propose a novel solution to it. In this study, we propose a new class of CAD models, i.e., Deep-Precognitive Diagnosis, wherein artificial agents are enabled to identify unknown diseases that have the potential to cause a pandemic in the future. A de novo biologically-inspired Conv-Fuzzy network is developed. Experimental results show that the model trained to classify Chest X-Ray (CXR) scans into normal and bacterial pneumonia detected a novel disease during testing, unseen by it in the training sample and confirmed to be COVID-19 later. The model is also tested on SARS-CoV-1 and MERS-CoV samples as unseen diseases and achieved state-of-the-art accuracy. The proposed model eliminates the need for model re-training by creating a new class in real-time for the detected novel disease, thus classifying it on all subsequent occurrences. Second, the model addresses the challenge of limited labeled data availability, which renders most supervised learning techniques ineffective and establishes that modified fuzzy classifiers can achieve high accuracy on image classification tasks.

Representative images showing Anteroposterior CXR scans of patients diagnosed with (a) Bacterial Pneumonia (b) SARS-CoV-1 (c) MERS-CoV (d) SARS-CoV-2.

Deep learning has revolutionized Computer-Aided Diagnosis (CAD), exhibiting remarkable diagnostic capabilities in complex clinical scenarios. However, conventional supervised DL models face limitations in recognizing novel diseases absent from the training dataset. Timely detection of emerging infectious diseases is crucial for effective control and prevention, especially when conventional models rely on post-outbreak data availability, as exemplified by the COVID-19 outbreak.

We introduced a novel approach, “Deep-Precognitive Diagnosis”, aimed at addressing the challenge of identifying unknown diseases with pandemic potential. A novel class of CAD models is proposed, featuring a biologically-inspired Conv-Fuzzy network. Unlike conventional models, these models enable artificial agents to detect novel diseases in real-time, eliminating the need for extensive re-training post-disease outbreak.

Experimental results demonstrate the efficacy of the proposed model in detecting a novel disease during testing, later identified as COVID-19, which was absent from the training dataset. Furthermore, the model exhibits state-of-the-art accuracy when tested on SARS-CoV-1 and MERS-CoV samples, showcasing its ability to generalize to unseen diseases. The model's capacity to create a new class in real-time for detected novel diseases obviates the need for continuous re-training. These results underscore the potential Deep-Precognitive diagnosis paradigm as a valuable tool in the precise diagnosis of novel diseases.

The proposed model not only addresses the limitation of conventional models in recognizing novel diseases but also tackles the challenge of limited labeled data availability. By leveraging modified fuzzy classifiers, the proposed model achieves high accuracy on image classification tasks, demonstrating its robustness in scenarios where supervised learning techniques fall short. Next steps for the project involve deploying the Deep-Precognitive diagnosis for a clinical trial and testing its performance over a wide survey sample of patients diagnosed with Pneumonia-type diseases.

The 3 categories of Neurons (a) Classifying Neuron (b) Overlap Compensation Neuron and (c) Containment Compensation Neuron, here f(x, y) and T(x) represents threshold functions.