| Abstract |
Smart healthcare system provides clinical care condition become more intelligence. It is developed using diverse sort of information technology components which include cloud computing, big data, Internet of Things (IoT), wearable devices, software application, image analysis, Artificial Intelligence (AI) algorithms and sensors. It penetrates into the multi-faceted aspects related to offering a rich set of medical services, effective tools for disease diagnosis, health monitoring, information pooling of patient records, identifying medical patterns, clinical decision making, forecasting medical trends or complications, drug discovery, recommend medical practices, delivering rehabilitation therapy, personalized and patient centric medication. The evolution of digital technologies, intelligent systems and smart devices empowered healthcare supports sustainable smart city development. In this work, a smart intelligent system is proposed for the staging classification of cervical cancer images by applying Automated Machine Learning (AutoML) using Google Cloud Platform (GCP). The description of the proposed system pipeline is given as follows. (1) Real time cervical cancer images from EVA digital colposcopy are collected from Kaggle site. These raw datasets are utilized to construct the proposed system, (2) The different stages of cervical cancer images are prepared, labelled and then uploaded into GCP environment, (3) A smart intelligent system is constructed using AutoML framework to categorize the stages of cervical cancer, and (4) The developed model is uploaded, stored and deployed in vertex Artificial Intelligence (vertex AI) platform to evaluate its accuracy. The outcome of this study demonstrates that the proposed system assists physicians to diagnose the stages of cervical cancer with minimal efforts and lesser time. It is significantly beneficial to healthcare professionals to arrive at clinical decisions without the involvement of taking biopsy and earlier detection of precancerous lesions. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023. |
