Malaria Parasite Detection using Efficient Neural Ensembles

Caused by the bite of the Anopheles mosquito infected with the parasite of genus Plasmodium, malaria has remained a major burden towards healthcare for years with an approximate 400,000 deaths reported globally every year. The traditional diagnosis process for malaria involves an examination of the blood smear slide under the microscope. This process is not only time consuming but also requires pathologists to be highly skilled in their work. Timely diagnosis and availability of robust diagnostic facilities and skilled laboratory technicians are very much vital to reduce the mortality rate. This study aims to build a robust system by applying deep learning techniques such as transfer learning and snapshot ensembling to automate the detection of the parasite in the thin blood smear images. All the models were evaluated against the following metrics - F1 score, Accuracy, Precision, Recall, Mathews Correlation Coefficient (MCC), Area Under the Receiver Operating Characteristics (AUC-ROC) and the Area under the Precision Recall curve (AUC-PR). The snapshot ensembling model created by combining the snapshots of the EfficientNet-B0 pre-trained model outperformed every other model achieving a f1 score - 99.37%, precision - 99.52% and recall - 99.23%. The results show the potential of model ensembles which combine the predictive power of multiple weal models to create a single efficient model that is better equipped to handle the real world data. The GradCAM experiment displayed the gradient activation maps of the last convolution layer to visually explicate where and what a model sees in an image to classify them into a particular class. The models in this study correctly activate the stained parasitic region of interest in the thin blood smear images. Such visuals make the model more transparent, explainable, and trustworthy which are very much essential for deploying AI based models in the healthcare network.

FinFET Towards Nano-Transistor: A Review

This paper review the FinFET structure as a future transistor for analog and digital electronic circuits, and present its electrical characteristics depending on the important parameters for evaluating the MOSFETs structures like DIBL and Ion/Ioff.

Design of an Electromyograph Equipped with Digital Neck Angle Elevation Gauge

The development of technology also affects human health, including body posture due to poor human position when using gadgets, both smartphones, and laptops. This study is design a tool that can measure the elevation of a person's neck angle equipped with electromyography, to help health workers, medical rehabilitation doctors to diagnose and provide treatment to patients with a bent head posture or forward head posture. In this research, an electromyography module is designed which consists of a series of instruments, a pre-amplifier circuit, a high pass filter, a low pass filter, and a dc offset regulator to be converted to digital so that it can be displayed on a laptop. In this study, the tapped muscle was the upper trapezius muscle using disposable electrodes. Meanwhile, to measure the angular elevation, the MPU 6050 sensor is used to measure the movement of the head forward. The frequency of the electromyography signal is 20-500 Hz. For software or display readings from this tool is Delphy. Meanwhile, the microcontroller used for ADC communication is Arduino Uno. From the research, it was found that the neck angle elevation gauge has a 0,597% error rate, for conditioning conducted on respondents, all respondents experienced a decrease in amplitude on the same frequency spectrum in the last ten minutes. Meanwhile, a drastic decrease occurred at the neck angle of 60°. Thus, it can be concluded that the forward position of the head affects the frequency spectrum of the neck muscles.

Malaria Parasite Detection using Efficient Neural Ensembles

Caused by the bite of the Anopheles mosquito infected with the parasite of genus Plasmodium, malaria has remained a major burden towards healthcare for years with an approximate 400,000 deaths reported globally every year. The traditional diagnosis process for malaria involves an examination of the blood smear slide under the microscope. This process is not only time consuming but also requires pathologists to be highly skilled in their work. Timely diagnosis and availability of robust diagnostic facilities and skilled laboratory technicians are very much vital to reduce the mortality rate. This study aims to build a robust system by applying deep learning techniques such as transfer learning and snapshot ensembling to automate the detection of the parasite in the thin blood smear images. All the models were evaluated against the following metrics - F1 score, Accuracy, Precision, Recall, Mathews Correlation Coefficient (MCC), Area Under the Receiver Operating Characteristics (AUC-ROC) and the Area under the Precision Recall curve (AUC-PR). The snapshot ensembling model created by combining the snapshots of the EfficientNet-B0 pre-trained model outperformed every other model achieving a f1 score - 99.37%, precision - 99.52% and recall - 99.23%. The results show the potential of model ensembles which combine the predictive power of multiple weal models to create a single efficient model that is better equipped to handle the real world data. The GradCAM experiment displayed the gradient activation maps of the last convolution layer to visually explicate where and what a model sees in an image to classify them into a particular class. The models in this study correctly activate the stained parasitic region of interest in the thin blood smear images. Such visuals make the model more transparent, explainable, and trustworthy which are very much essential for deploying AI based models in the healthcare network.

A Smart System to Ease Occurrence of Bedsores

The occurrence of bedsores in Ghanaian hospitals, elderly homes or care homes is especially high among patients or people who are incapacitated and cannot move or turn on their own, and who happen to remain in a particular posture for a very long time. Patients in coma, those operated on and for that matter in critical state, and patients confined to wheelchairs are primary examples. Constant pressure on some parts of the body leads to the occurrence of pressure sores or ulcers. This paper seeks to implement a Bedsore Easing System (BeSoSys) that integrate several embedded hardware components, a database and software to reduce the occurrence of bedsores. These embedded hardware components include the Bed Device Unit (BDU), the Pocket Device Unit (PDU), a pressure or weight sensor, a temperature sensor, and an inflation-deflation device. The BDU is fitted into the bed of the patient or on the surface of skin contact of the patient. The PDU is assigned to nurses or caretakers to serve as an alarm system for patient repositioning depending on situation. All activities in the Bedsore Easing System are logged into a database for future references. A bedridden patient exerts constant pressure on the bony protrusions of the body, and this causes bedsores. It was found out during the research that in Ghana, the nurses or caretakers used to turn and massage patients at some random time intervals as a way of preventing bedsores. This traditional way of turning and massaging patients is not only tedious but also ineffective. This paper seeks to provide easy, better, and effective solution to ease bedsores. The BeSoSys intends to prevent the occurrence of bedsores hence the alleviation of bedsore complications

A Two Channels Wireless Electrocardiograph System Using Bluetooth Communication

Health problems with cardiovascular system disorders are still ranked high, according to data from the WHO reported that there are about 31% of causes of death globally are cardiovascular diseases. The purpose of this study was to develop a 12 lead electrocardiograph with 2 displays and the HC-05 as a real-time transmitter of heart signal data. The electrocardiogram signal is obtained from the wiretapping by attaching the electrode cable to the Lead I, Lead II, Lead III, aVR, aVL, and aVF leads, then processed on IC AD620, HPF and LPF filters and non-inverting amplifiers and then processed using Arduino UNO for further display. in the form of a signal on the Delphi 7 application. The research method is to measure the heart signal on the ECG Simulator, by testing several BPMs, namely 30, 60, 120 and 240 on each lead. After testing the signal equation at the 0.5mV setting by calculating the error rate, the highest error value is obtained in lead I, lead aVL and aVF of 7.14% and the smallest error is 3.57% error in lead III. Then at the 1mV setting by calculating the error rate, the highest error value in lead aVL is 7.14% and the smallest error is 2.36%. at the 2mV setting by calculating the error rate, the highest error value is obtained in leads aVL and aVF of 5.71% and the smallest error is obtained by an error of 2.1% in lead II. the results of this study are implemented so that in the future an ECG examination can be carried out and then monitored remotely like a doctor's room because the data communication uses bluetooth.

A comparative Evaluation of Swarm Intelligence Algorithm Optimization: A Review

Swarm intelligence (SI), an important aspect of artificial intelligence, is increasingly gaining popularity as more and more high-complexity challenges necessitate solutions that are sub-optimal but still feasible in a fair amount of time. Artificial intelligence that mimics the collective behavior of a group of animals is known as swarm intelligence. Attempting to survive. It is primarily influenced by biological systems. The main aim of our article is to find out more about the guiding principle, classify possible implementation areas, and include a thorough analysis of several SI algorithms. Swarms can be observed in ant colonies, fish schools, bird flocks, among other fields. During this article, we will look at some Swarm instances and their behavior. We see many Swarm Intelligence systems, like Ant colony Optimization, which explains ant activity, nature, and how they conquer challenges; in birds, we see Particle Swarm Optimization is a swarm intelligence-based optimization technique, and how the locations must be positioned based on the three concepts. The Bee Colony Optimization follows, and explores the behavior of bees, their relationships, as well as movement and how they work in a swarm. This paper explores some of the methods and algorithms.

Apnea Monitor Using Pulse Oxymetry with Tactile Stimulation to Reduce Respiration Failure

Respiratory failure (apnea) often occurs in premature babies, this should be avoided because it causes low oxygen concentrations in the blood so that it can damage brain function and lead to death. Apnea is characterized by a decrease in oxygen saturation (SpO2). The purpose of this study was to design an apnea monitor that was detected with SpO2 parameters, alarms, and vibrating stimulation. This study uses infrared and red LEDs that emit light through the surface of the finger and is detected by a photodiode sensor, this light signal will be converted into an electrical signal and calculated by Arduino to determine the patient's SpO2 and BPM values. If the SpO2 value drops 5% within 5 seconds from the baseline, the device will indicate apnea has occurred and the vibrating motor is working. SpO2 signals and alarms are sent to the nurse station computer via Bluetooth HC-05. The instrument was calibrated with an SpO2 calibrator and the measurement results were compared with a BION pulse oximetry brand. The results of the instrument measurement on two subjects on the SpO2 parameter showed an error value of 2% and the BPM parameter obtained an error value of 4.54%. Testing the BPM parameter using a calibrator at the 30 and 60 BPM settings shows an error value of 0% and at the 120 BPM setting the error value is 0.01%. The vibrating motor to stimulate the baby's body when apnea occurs is functioning properly. The results showed that measurements using subjects tended to have high error values ​​due to several factors. This research can be implemented on patient monitors to improve patient safety and reduce the workload of nurses or doctors

Oximeter and BPM on Smartwatch Device Using Mit-App Android with Abnormality Alarm

Monitoring is an activity that is carried out continuously. Healthy condition is a parameter that is needed in life, one of the important parameters is the measurement of oxygen saturation in the blood and heart rate. The purpose of this research is to develop a Smartwatch SpO2 device and BPM sensor that is connected to WIFI using the Android Platform instead of using an LCD for parameter reading. This module design method uses the MAX30100 sensor to display the SpO2 and BPM values ​​displayed on the OLED. Data processing is carried out using ATMEGA 328P programming and then displayed in the Android-based Mit-app application. The results show the average error for the SPO2 value is 0.868 % and the standard deviation is 0.170 %, while the BPM value has an average error of 0.56 % and a standard deviation of 0.30%. From the results of the comparison data analysis, the largest error was 1.03% and the smallest was 0.62% for Spo2 ml/hour with an accuracy of 0.05 (0.57%) with a precision value of 0.08 at the selection speed of 50 ml/hour. From the results above, it can be concluded that the data can be displayed on OLED using the Mit-app Android application with an error rate accuracy of 0.57%. From the results of this research design, it is hoped that it can facilitate the diagnosis of the condition of patients and health nurses

Performance Evaluation of IoT-based SpO2 Monitoring Systems for COVID-19 Patients

Internet of Things (IoT) applications can be used in healthcare services to monitor patients remotely. One implementation is that it is used to monitor COVID-19 patients. During the COVID-19 pandemic, people who are infected without symptoms must self-isolate so that the virus does not spread. Measurement of blood oxygen levels or SpO2 is one of the measurements that must be carried out in routine examination procedures for self-isolating patients for early detection of silent hypoxemia in COVID-19 patients. Previous research has developed an IoT-based health monitoring system with a Wireless Body Sensor Network (WBSN) and a gateway that can be used for data acquisition and transmission. The system uses a home pulse oximeter to measure SpO2 and heart rate and an Android application that functions as an IoT gateway to collect data from sensors and add location information before sending data to the server. The WBSN has been successfully integrated with two types of open source IoT platforms, namely ThingsBoard and Elasticsearch Logstash Kibana (ELK). However, it is necessary to carry out further studies on analytical and experimental performance tests of the two systems. Therefore, the purpose of this study is to develop a performance evaluation of the IoT-based SpO2 monitoring systems using the Thingsboard and ELK as IoT platforms. To evaluate the performace we ran the monitoring system on both platforms using pulse oximeter and Android device as IoT gateway with HTTP and MQTT as transport protocol for sending the data to the server. From this study we found that average time of message delivery in ELK compared to ThingsBoard using the same protocols was higher but stable.