The emergence of Cloud Healthcare Platform for Colorectal Cancer (CC) diagnosis is crucial for effective detection and management of CC, a cancer that originates in the colon or rectum. Timely CC diagnosis considerably increases the possibilities of patient survival and successful treatment. Automated CC recognition systems are possible to enhance early analysis and treatment outcomes, decrease medical bills, and enhance access to diagnostic and screening services. Computer-aided diagnoses (CAD) system analyses clinical images like colonoscopy or CT scans, for detecting suspicious lesions, polyps, or tumors in the colon and rectum. This cloud-based healthcare platform employs advanced computer-based technology and system to facilitate accurate and early CC diagnosis through the analysis of medical images and patient information. The main objective is to improve the efficiency and reliability of CC screening, monitoring, and diagnoses. Deep learning (DL) approaches, particularly convolutional neural networks (CNNs), has shown promise in automatic CC diagnosis and are leveraged in this platform. Therefore, this study presents a new Black Widow Optimization with Deep Ensemble Voting Classifier (BWO-DEVC) technique for CC detection in the cloud platform. The objective of BWO-DEVC method is to recognize and classify the occurrence of the CC on medical images using the cloud platform. At the initial stage, the BWO-DEVC technique involves storing the medical images into the cloud environment where the execution process will be performed. In addition, the BWO-DEVC technique follows DarkNet-53 feature extractor to generate a set of feature vectors. For CC classification, the BWO-DEVC technique follows ensemble voting classifier encompassing three DL algorithms like gated recurrent unit (GRU), bidirectional LSTM (BiLSTM), and long shortterm memory (LSTM). Finally, the hyperparameter selection of the DL models takes place using the BWO algorithm, which in turn enhances the CC detection results. An extensive set of experiments were made to validate the enriched CC detection results of the BWO-DEVC algorithm. The extensive outcomes highlighted that the BWO-DEVC method reaches high performance than other techniques in the CC diagnostic process.