Conv-MTD: A CNN Based Multi-Label Medical Tubes Detection and Classification Model to Facilitate Resource-Constrained...

Computer-aided detection through deep learning is becoming a prevalent approach across various fields, including the detection of anomalies in medical procedures. One such medical procedure involves the placement of medical tubes to provide nutrition or other medical interventions in critically ill patients. Medical tube placement can be highly complex and prone to subjective errors. Malposition of medical tubes is often observed and associated with significant morbidity and mortality. In addition, continuous verification using manual procedures such as capnography, pH testing, auscultation, and visual inspection through chest X-ray (CXR) imaging is required. In this paper, we propose a Conv-MTD, a medical tube detection (MTD) model that detects the placement of medical tubes using CXR images, assisting radiologists with precise identification and categorizing the tubes into normal, abnormal, and borderline placement. Conv-MTD leverages the EfficientNet-B7 architecture as its backbone, enhanced with an auxiliary head in the intermediate layers to mitigate vanishing gradient issues common in deep neural networks. The Conv-MTD is further optimized using post-training 16-bit floating-point (FP16) quantization, which significantly reduces memory consumption by 50% and 2x improvement in inference speed without compromising accuracy. This optimization allows Conv-MTD to achieve efficient performance without requiring high-end computational resources, making it suitable for deployment on point-of-care devices. Conv-MTD provided the best performance, with an average area under the receiver operating characteristic curve (AUC) of 0.95 using the open-source RANZCR CLiP dataset. The proposed Conv-MTD has the potential to operate on resource-constrained point-of-care devices due to its use of FP16 computation, enabling low-cost and automated assessments in various healthcare settings.