Kok Beng Gan

Comparison of adaptive neuro-fuzzy inference system and artificial neutral networks model to categorize patients in the emergency department

Unexpected disease outbreaks and disasters are becoming primary issues facing our world. The first points of contact either at the disaster scenes or emergency department exposed the frontline workers and medical physicians to the risk of infections. Therefore, there is a persuasive demand for the integration and exploitation of heterogeneous biomedical information to improve clinical practice, medical research and point of care. In this paper, a primary triage model was designed using two different methods: an adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN).When the patient is presented at the triage counter, the system will capture their vital signs and chief complains beside physiology stat and general appearance of the patient. This data will be managed and analyzed in the data server and the patient’s emergency status will be reported immediately. The proposed method will help to reduce the queue time at the triage counter and the emergency physician’s burden especially duringdisease outbreak and serious disaster. The models have been built with 2223 data set extracted from the Emergency Department of the Universiti Kebangsaan Malaysia Medical Centre to predict the primary triage category. Multilayer feed forward with one hidden layer having 12 neurons has been used for the ANN architecture. Fuzzy subtractive clustering has been used to find the fuzzy rules for the ANFIS model. The results showed that the RMSE, %RME and the accuracy which evaluated by measuring specificity and sensitivity for binary classificationof the training data were 0.14, 5.7 and 99 respectively for the ANN model and 0.85, 32.00 and 96.00 respectively for the ANFIS model. As for unseen data the root mean square error, percentage the root mean square error and the accuracy for ANN is 0.18, 7.16 and 96.7 respectively, 1.30, 49.84 and 94 respectively for ANFIS model. The ANN model was performed better for both training and unseen data than ANFIS model in term of generalization. It was therefore chosen as the technique to develop the primary triage prediction model. This primary triage model will be combined with the secondary triage prediction model to produce the final triage category as a tool to assist the medical officer in the emergency department.

Transabdominal Fetal Heart Rate Detection Using NIR Photopleythysmography: Instrumentation and Clinical Results

In obstetrics, fetal heart rate (FHR) detection remains the standard for intrapartum assessment of fetal well-being. In this paper, a low-power (<55 mW) optical technique is proposed for transabdominal FHR detection using near-infrared photoplesthysmography (PPG). A beam of IR-LED (890 nm) propagates through to the maternal abdomen and fetal tissues, resulting in a mixed signal detected by a low-noise detector situated at a distance of 4 cm. Low-noise amplification and 24-bit analog-to-digital converter resolution ensure minimum effect of quantization noise. After synchronous detection, the mixed signal is processed by an adaptive filter to extract the fetal signal, whereas the PPG from the mothers index finger is the reference input. A total of 24 datasets were acquired from six subjects at 37 plusmn 2 gestational weeks. Results show a correlation coefficient of 0.96 (p-value < 0.001) between the proposed optical and ultrasound FHR, with a maximum error of 4%. Assessment of the effect of probe position on detection accuracy indicates that the probe should be close to fetal tissues, but not necessarily restricted to head or buttocks.

Initialization of adaptive neuro-fuzzy inference system using fuzzy clustering in predicting primary triage category

This paper describes the fuzzy clustering method to initialize the Adaptive Neuro-Fuzzy Inference System (ANFIS) in predicting primary triage category. Fuzzy C-means (FCM) and Fuzzy Subtractive clustering (FSC) are the most commonly used unsupervised clustering methods to initialize the ANFIS model. A total of 135 data was extracted from Objective Primary Triage Scale (OPTS) records obtained from Emergency Department UKMMC. These data was used to develop the ANFIS model and predict the primary triage category. The classification accuracy of the ANFIS model using fuzzy clustering method in predicting the primary triage category is 98.4%. The FCM method produced fewer rules and needed less processing time to reach the RMSE of 0.127 compared to the FSC method. These results suggest that FCM clustering will be used to predict the primary triage category.

Design and Implementation of A PC-Based Digital Synchronous Detection System for Biological Signal Measurement

A PC-based digital lock-in amplifier was implemented completely in software (Labview) for photoplethysmograph (PPG) measurement. Experiments carried out to evaluate the performance of the amplifier show that the system can recover the signals up to 40 dB below the interference (ambient light). This system can be customized later to measure the trans-abdominal PPG signal of pregnant women for fetal heart rate detection.

Feasibility of Fetal Photoplethysmography Signal Extraction using Adaptive Noise Cancelling

In this paper, an approach based on adaptive noise cancellation (ANC) is evaluated to extract the fetal heartrate using photoplethysmographic (PPG) signals from the maternal abdomen. Results of simulations using semisynthetic PPG signals are presented which show the feasibility of the proposed technique. Then a mixture of PPG signal is produced by recording the PPG from overlapping fingers from two different subjects in-lieu of the mother and fetus. Results show that a recursive least-squares algorithm is capable to extract the peaks of the desired PPG signal, hence the heartrate, even at a SNR of -34 dB.

Secondary triage classification using an ensemble random forest technique

BACKGROUND Triage of patients in the emergency department is a complex task based on several uncertainties and ambiguous information. Triage must be implemented within two to five minutes to avoid potential fatality and increased waiting time. OBJECTIVE An intelligent triage system has been proposed for use in a triage environment to reduce human error. METHODS This system was developed based on the objective primary triage scale (OPTS) that is currently used in the Universiti Kebangsaan Malaysia Medical Center. Both primary and secondary triage models are required to develop this system. The primary triage model has been reported previously; this work focused on secondary triage modelling using an ensemble random forest technique. The randomized resampling method was proposed to balance the data unbalance prior to model development. RESULTS The results showed that the 300% resampling gave a low out-of-bag error of 0.02 compared to 0.37 without pre-processing. This model has a sensitivity and specificity of 0.98 and 0.89, respectively, for the unseen data. CONCLUSION With this combination, the random forest reduces the variance, and the randomized resembling reduces the bias, leading to the reduced out-of-bag error.

Development of a computer-based automated pure tone hearing screening device: a preliminary clinical trial

Abstract Hearing screening is important for the early detection of hearing loss. The requirements of specialized equipment, skilled personnel, and quiet environments for valid screening results limit its application in schools and health clinics. This study aimed to develop an automated hearing screening kit (auto-kit) with the capability of realtime noise level monitoring to ensure that the screening is performed in an environment that conforms to the standard. The auto-kit consists of a laptop, a 24-bit resolution sound card, headphones, a microphone, and a graphical user interface, which is calibrated according to the American National Standards Institute S3.6-2004 standard. The auto-kit can present four test tones (500, 1000, 2000, and 4000 Hz) at 25 or 40 dB HL screening cut-off level. The clinical results at 40 dB HL screening cut-off level showed that the auto-kit has a sensitivity of 92.5% and a specificity of 75.0%. Because the 500 Hz test tone is not included in the standard hearing screening procedure, it can be excluded from the auto-kit test procedure. The exclusion of 500 Hz test tone improved the specificity of the auto-kit from 75.0% to 92.3%, which suggests that the auto-kit could be a valid hearing screening device. In conclusion, the auto-kit may be a valuable hearing screening tool, especially in countries where resources are limited.

Determination of Reflectance Optical Sensor Array Configuration Using 3-Layer Tissue Model and Monte Carlo Simulation

A new reflectance optical sensor array for locating fetal signal transabdominally has been determined in this study. The selection of optical sensor array is based on the highest Irradiance (μW/m2) value estimated on respected detectors. A three-layer semi-infinite tissue model which consists of maternal, amniotic fluid sac and fetal tissues is employed to study the optical sensor array configuration. By using statistical error approach, the number of rays injected to the system can be set to 1 million rays with ±3.2% of simulation error. The simulation results obtained from Monte Carlo technique reveal that diamond configuration is the most suitable configuration of reflectance optical sensor array with 40mm of emitter-detector separation. The selected configuration will be useful in detecting fetal signal independently of probe position.

Application of Adaptive Noise Cancellation in Transabdominal Fetal Heart Rate Detection Using Photoplethysmography

Monitoring the fetal heart rate (FHR) throughout pregnancy empowers the clinician to diagnose fetal well being, characterize fetal development and detect abnormality (Freeman et al. 2003). A non-invasive and low cost system would enable monitoring of normal pregnancies and promote large population studies of fetal physiological development (Freeman et al. 2003). FHR monitoring is an ongoing observation of human fetal physiology. The expected outcome of this early detection is a reduced risk of fetal morbidity and mortality (Philip et al. 2002). Currently, Doppler ultrasound has been extensively used for FHR detection and obstetric purposes (Hershkovitz et al. 2002), where the standard pre-delivery test of fetal health is the fetal non-stress test (NST). These tests are routinely performed at the hospital where continuous-wave instruments are more popular than pulsed ones. The use of Doppler ultrasound in the first trimester is generally not recommended as a routine (Hershkovitz et al. 2002) as it may increases the occurrence of intrauterine growth restriction. Besides that, the FHR measurements using Doppler ultrasound are not always reliable (Karlsson et al. 2000) due to the complexity of the Doppler signal and the effects of fetal and maternal breathing. An alternative to ultrasound is using the fetal electrocardiogram (FECG). In direct (invasive) fetal electrocardiogram (FECG), the FHR could be obtained by attaching scalp electrode to the fetal scalp after the rupture of the membrane (Khandpur 2004). During invasive FECG recording the uterus may be perforated leading to its infection, besides possible scalp injuries to the fetus (Khandpur 2004). The other approach is non-invasive FECG but FECG signals have a low (signal to noise ratio) SNR due to the interference from noise, maternal electrocardiogram (MECG) and electromyogram (EMG). The application of non-invasive FECG requires multiple leads and advanced digital signal processing techniques (Najafabadi et al. 2006). It is worth mentioning that commercial devices operating on non-invasive FECG are not available at this moment. Optical techniques has received a considerable attention in biomedical diagnostic and monitoring of biological tissues such as brain imaging, breast imaging and for fetal heart rate detection and oxygen saturation measurement due to its theoretical advantages in

Two channel abdominal PPG instrumentation

In this paper, a two channel abdominal PPG instrumentation is developed. The proposed instrument consists of IR-LED and its driver, photo-detector and data acquisition card. The modulation frequency generation, demodulation and digital signal processing is done completely in the digital domain using LabView. The results show that the developed instrument is able to acquire signal from the abdomen even at 4 cm source to detector separation. This instrument is intended for future application in trans-abdominal fetal heart rate detection.