Automatic Modulation Recognition with Breast Cancer Classification for Medical Internet of Things Applications

• Published in: 2023 International Conference on Emerging Research in Computational Science (ICERCS) pp. 1 - 6
• Main Authors: Nirmalrani, V., Poornima, D., Gowri, S., Chandrasekaran, Kumaresan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 07.12.2023
• Subjects: Automatic Modulation Recognition; Breast cancer; Deep belief networks; Internet of Things; Medical diagnostic imaging; Medical Internet of Things; Modulation; Sensitivity; Wireless communication; Wireless sensor networks
• DOI: 10.1109/ICERCS57948.2023.10434250

Medical Internet of Things (MIoT) and healthcare have lately converged, which has led to novel techniques to identifying and treating medical troubles. This research gives a completely unique method that utilizes the Internet of Things (IoT) to combine Automatic Modulation Recognition (AMR) with Breast Cancer (BC) categorization. The loss of included answers arises from the reality that most current research ignores BC type in desire for both AMR or BC type. A unique Medical Internet of Things-based totally Breast Cancer (MIoT-BC) category method addresses the project. In order to facilitate automatic modulation identity, the proposed gadget first statistics indicators from clinical gadget and wireless verbal exchange channels. After that, it uses DBNs (deep belief networks) to categorize BCs. The MIoT-primarily based strategy ensures the secure transmission and instantaneous analysis of scientific facts, which improves the precision and timeliness of diagnosis. There are some of blessings to the use of MIoT for BC categorization, such as less difficult get entry to to health information, faster diagnostic instances, and the option for faraway monitoring. In addition, the potential for misclassification and human errors is reduced, making this a useful tool inside the realm of medical diagnostics. DBNs are used to give a dependable and correct type version for BC categorization. Several healthcare packages have found achievement with DBNs due to their suitability for processing excessive-dimensional, complicated scientific facts. Differentiating benign from malignant breast lesions is an area where their capability to extract sizeable facts from enter facts robotically shines. MIoT-primarily based processes the usage of DBNs enhance BC classification accuracy in comparison to traditional techniques, consistent with this paper. Evaluation criteria, such as sensitivity, specificity, accuracy, overall performance and F1 rating, measure the effectiveness of the suggested machine.