A goniometer measures the available range of motion of an object. In medicine, this instrument is used assessment tool for joints in the body that allows physicians to evaluate a patient's condition, which then enables therapeutic adjustments. In order to improve recording diagnosis history, and the measurement process itself, this study aimed to create a digital goniometer that allows instantaneous measurement of the elbow and knee joint angles, through pictures. The device was made possible through the use of Raspberry Pi, a microcomputer capable of supporting functional applications through integration of hardware and software components. Statistical analysis shows a 98.25% and 98.09% accuracy for the elbow and knee joints respectively, resulting from closely related values between the actual goniometer, and the device created from this study.

Image-Processing-based Digital Goniometer using OpenCV

Since many diseases and infections are dependent on the count and type of Red Blood Cells (RBCs) and White Blood Cells (WBCs) present in the blood stream, detection and classification pertaining to them is necessary and relevant. Based from existing related literature, ordinary Neural Networks are usually employed. Also, in existing researches, RBC types are the main focus. Hence, after observing research gaps, a Faster Region-based Convolutional Neural Network (Faster R-CNN) was utilized for this study, focusing not only on RBCs but also on the variants of WBCs. The aim is to have a fast and reliable system in order to achieve the goal of aiding the medical field in the classification of RBCs and WBCs.

Faster R-CNN Model With Momentum Optimizer for RBC and WBC Variants Classification

The differential count of white blood cells (WBCs) is one widely used approach to assess the status of a patient's immune system. Currently, the main methods of differential WBC counting are manual counting and automatic instrument analysis with labeling preprocessing. But these two methods are complicated to operate and may interfere with the physiological states of cells. Therefore, we propose a deep learning-based method to perform label-free classification of three types of WBCs based on their morphologies to judge the activated or inactivated neutrophils. Over 90% accuracy was finally achieved by a pre-trained fine-tuning Resnet-50 network. This deep learning-based method for label-free WBC classification can tackle the problem of complex instrumental operation and interference of fluorescent labeling to the physiological states of the cells, which is promising for future point-of-care applications.

https://www.mdpi.com/1424-8220/21/2/512/pdf

Deep-Learning Based Label-Free Classification of Activated and Inactivated Neutrophils for Rapid Immune State Monitoring.

Biomedical image analysis by machine learning enables accurate cell counting and localization, which are essential to clinic diagnosis and research. However, high-density population of the cells poses great challenges. The variability and intricacy in the shapes and colours of cells further jeopardize the performance of existing models. Aiming to overcome the difficulties, we propose an end-to-end manifold-regularized regression network (MRRN) for cell counting and localization. The backbone is a U-Net tweaked for regression and trained by density maps to preserve the spatial distribution of the cells. Based on graph Laplacian, the patch-wise manifold regularizer imposed on the U-Net can represent the intrinsically nonlinear and complex features of the images, thus increasing the generalizability of the network and lowering the risk of overfitting. Experimental results show that MRRN outperforms several state-of-the-art algorithms on a widely-adopted synthetic data set and a real-world carcinoma data set, which verifies the efficacy and applicability of the proposed model.

Manifold-Regularized Regression Network: A Novel End-to-End Method for Cell Counting and Localization

/pdf/classification-of-white-blood-cells-using-convolutional-1szhw6iwhc.pdf

Classification of White Blood Cells Using Convolutional Neural Network

Blood cells both white and red are important part of the immune system. These cells help fight infections by attacking bacteria, viruses, and germs that invade the body. White blood cells originate in the bone marrow but circulate throughout the bloodstream, while red blood cell helps transport oxygen to our body. Accurate counting of those may require laboratory testing procedure that is not usual to everyone. Generating codes that will help counting of blood cells that produce accurate response via images gives a relief on this problem. In this study, the images were processed and a blob detection algorithm was used to detect and differentiate RBCs from WBCs. A cell counting method was also used to provide an actual count of the RBCs and WBCs detected. The automation comes with a graphical user interface backed-up with a working database system to keep the records of the users (e.g. patients, respondents). The performance of the system was statistically described as accurate compared to the manual method of counting. Results show an accuracy of 100% for platelet, 96.32% for RBCs and 98.5% for WBCs. Hence, the proposed system can benchmark with the manual methods of detection and counting of platelets, RBCs and WBCs in blood samples.

Blood Cells Counting using Python OpenCV

A design of a microcontroller-based wheelchair controlled by the motion of the eye is presented. The user wears a set of electronic eyeglasses used to detect the movement of the sclera and then transmits control signals to the wheelchair via Bluetooth. Detection of sclera movement is done via infrared sensors. The wheelchair is equipped with an obstacle detection system utilizing ultrasonic sensors. To operate the wheelchair, the user presses a button to move forward, and moves his eyes to turn the chair in a specific direction. The sensors can detect four directions of eye movement: half left, full left, half right and full right. Tests for response to eye movement, delay in the transmission of control signals, reliability of transmission and reliability of the ultrasonic sensors in terms of obstacle detection shows that the design met the objectives to ensure proper and safe operation of the wheelchair.

Eye-Controlled Wheelchair Using Infrared Detection

The main purpose of this project is to design and develop a software based computer program that will convert Filipino Speech into its equivalent Sign Language. This program will run in the personal computer through MATLAB and will use Hidden Markov Model (HMM) as speech recognition engine that uses a MFCC coefficient in improving the speech feature representation in a HMM based recognition speech. It uses 7 state left-to-right model and a word-based model for a small scale vocabulary speech recognition. This study involves testing the system’s speech recognition reliability and accuracy in which it is indicated in the objectives of the project. The analysis is expected to meet the 85% reliability of the speech recognition of the system proposed in the objective. Keyword MATLAB, Speech to Sign Language, Accuracy, Speech recognition, HMM (Hidden Markov Model).

Development of Speech to Filipino Sign Language for the Deaf or Mute Filipinos Using Hidden Markov Model (HMM)

Closed-circuit television camera (CCTV) and thermal imaging devices are used to detect febrile individuals entering establishments for Coronavirus 2019 (COVID-19) containment. Real-time tracking in post-COVID is manually checked by security personnel, which has risks of less efficiency due to human errors, as advance thermal cameras are unaffordable for some business owners. The main goal is converting an installed CCTV interfaced with infrared sensor to develop an economical thermal screening system with acoustic alarm. In this project, the colored and heatmap images transmitted from the thermal camera were processed through OpenCV. A calibration method was also performed to validate the temperature reading from the thermal camera. The project comes with graphical user interface (GUI) connected into a database, which visually tracks individuals exhibits elevated body temperature. The performance of the system shows above 95% accuracy upon conducting an inexpensive calibration check. The significance of this project is highlighting the effective mitigation of virus spread which offers safe and contactless analysis of potential individuals showing early symptoms of COVID-19. Additional features can be added for future work such as facemask detector, multiple thermal camera setup, and Login Options making the device and application exclusively for business owners.

Thermal Screening Solution Using Infrared Thermography and Closed-Circuit Television

This study improved upon a previous robot that could detect the agents of Triacetone triperoxide (TATP), which used an electronic nose detection system equipped with gas sensors to replace dogs used in detecting non-nitrogen-based precursor chemical compounds in Improvised Explosive Devices. The newly designed prototype had a 2-kilogram lighter build, had 4.44 watt-hours less robot power consumption and 7.4 watt-hours less controller power consumption; had better mobility, travelling on different surfaces with a maximum inclination of 45 degrees and climb 3-inch steps within a line-of-sight radius of 41 to 45 meters from the end-user; and could transmit and SMS alert to a selected authority. The new system also detected Potassium chlorate, Potassium perchlorate, and Sodium chlorate, with 90% overall accuracy, and classify all stated chemicals with an overall sensitivity and specificity of 57% and 88.4% respectively when operated within a line-of-sight radius of21 to 25 meters and within 5 seconds when the controller’s battery is above 60%. The system easily detected 2 chemicals inside the plastic, fabric, and paper containers but had difficulty detecting the other 3 in the varying containers. The use of medical-grade sensors, a wider array of sensors, or both, may alleviate issues found.

Raymond Joseph Meimban

Papers

Blood Cells Counting using Python OpenCV

Eye-Controlled Wheelchair Using Infrared Detection

Development of Speech to Filipino Sign Language for the Deaf or Mute Filipinos Using Hidden Markov Model (HMM)

Thermal Screening Solution Using Infrared Thermography and Closed-Circuit Television

Development of a Remote-Controlled Non-Nitrogen-Based IED Detecting Robot using an Electronic Nose System