Patient monitoring is an essential practice in healthcare settings to assess well-being and medical conditions. This study proposes a central patient monitoring system enabling healthcare personnel to have a comprehensive perspective of multiple patients' health state from a central location. The proposed prototype enables healthcare providers to detect and respond to changes in patients' condition efficiently. The project contains multiple patient monitoring units with each measuring temperature, ECG, SpO2, and heart rate along with a display on a centralized screen. The hardware of monitors is placed at each patient bed connected with all the patients and the result of each patient can be seen from the general ward counter. Different sensors are used to design monitoring units and are connected with node MCU through different pin connections. Node MCU sends the information gathered from sensors to the Raspberry pi unit using Wi-Fi. The real-time data of patients are being displayed and saved using server “thing speak”. There are different options on the server window to switch between different patients’ real time results to monitor them. The proposed prototype uses computer technology to monitor the vital signs of multiple patients on a single monitor from a centralized location using internet which may eventually reduce the time and workload of medical staff leading to improved patient care.

Membrane proteins constitute essential biomolecules attached to or integrated into cellular and organelle membranes, playing diverse roles in cellular processes. Their precise localization is crucial for understanding their functions. Existing protein subcellular localization predictors are predominantly trained on globular proteins; their performance diminishes for membrane proteins, explicitly via deep learning models. To address this challenge, the proposed study segregates membrane proteins into three distinct locations, including the plasma membrane, internal membrane, and membrane of the organelle, using deep learning algorithms including recurrent neural networks (RNN) and Long Short-Term Memory (LSTM). A redundancy-curtailed dataset of 3000 proteins from the MemLoci approach is selected for the investigation, along with incorporating pseudo amino acid composition (PseAAC). PseAAC is an exemplary technique for extracting protein information hidden in the amino acid sequences. After extensive testing, the results show that the accuracy for LSTM and RNN is 83.4% and 80.5%, respectively. The results show that the LSTM model outperforms the RNN and is most commonly employed in proteomics.