Claim Missing Document
Check
Articles

Classification of CT Scan Images of Stroke Patients and Normal Brain Based on Histogram, GLCM, and GLRLM Texture Features using K-Nearest Neighbor Azizah, Fitria Kholbi; Putri, Diana Salsabila; Permana, Riyan; Sumarti, Heni; Darma, Panji Nursetia
Journal of Physics and Its Applications Vol 7, No 4 (2025): November 2025
Publisher : Diponegoro University Semarang Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/jpa.v7i4.27259

Abstract

Stroke is a major neurological disorder requiring rapid and accurate diagnosis for effective treatment. Computerized Tomography (CT) scanning provides detailed brain imaging but requires expert interpretation. This study aims to develop an automated classification system to distinguish between normal and stroke-affected brain CT scan images using texture feature analysis, providing enhanced accuracy and robustness compared to existing single-feature approaches. A total of 200 CT scan images (100 normal, 100 stroke cases) from the Kaggle database were analyzed. Texture features were extracted using Histogram, Gray Level Co-occurrence Matrix (GLCM), and Gray Level Run Length Matrix (GLRLM) analysis. The KNN algorithm was evaluated using percentage split validation, with the training set ranging from 50% to 70% of the data. The KNN classifier achieved optimal performance with 93% accuracy, 91% precision, and 96% recall using a 50% training set, demonstrating its potential as a diagnostic support tool for healthcare professionals to facilitate faster diagnosis and treatment decisions. The integration of multiple texture analysis methods showed superior performance compared to individual feature extraction techniques. Histogram features contributed significantly to classification accuracy by enhancing the detection of tissue heterogeneity. Texture analysis revealed significant differences between normal and stroke images in entropy, contrast, and correlation parameters. The proposed method successfully classifies CT scan images of normal and stroke-affected brains with high accuracy, demonstrating potential for clinical implementation in automated stroke screening and diagnostic support.
Pilot Study: Portable Non-Invasive Blood Sugar, Cholesterol, Uric Acid Monitoring System Sumarti, Heni; Alvania Nabila Tasyakuranti; Qolby Sabrina
Jurnal Teknik Elektro Vol. 16 No. 1 (2024)
Publisher : LPPM Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/jte.v16i1.8204

Abstract

Degenerative diseases commonly associated with abnormal blood sugar, cholesterol, and uric acid levels require regular monitoring. Remote health monitoring technology enables children to monitor their parents' health conditions from a distance. This research presents a prototype development through Research and Development (R&D) methodology. This study developed a portable, low-cost, non-invasive detection system for blood sugar, cholesterol, and uric acid levels using the TCRT5000 sensor with Telegram integration. The compact device offers real-time monitoring advantages without blood sampling. The development followed the ADDIE (Analysis, Design, Development, Implementation, and Evaluation) model. The research results show the prototype's coefficient of determination for blood sugar is 0.9733, cholesterol is 0.9411, and uric acid is 0.9610. The non-invasive prototype demonstrates measurement errors of 7.41% for blood sugar, 15.83% for cholesterol, and 14.69% for uric acid. These error rates currently exceed medical measurement standards. The system successfully integrates with the Telegram application for remote monitoring. Future research should incorporate artificial intelligence algorithms to minimize error values.
Portable EMG System for IoT Using MQTT Yuniati, Anis; Ramadhani, Bintang; Rakhmadi, Frida Agung; Sumarti, Heni
Jurnal Teknik Elektro Vol. 17 No. 2 (2025)
Publisher : LPPM Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/jte.v17i2.33839

Abstract

Conventional electromyography (EMG) systems in clinical settings present limitations including high examination costs (500,000-1,500,000 IDR), bulky equipment, and lack of remote monitoring capabilities. This study develops and validates a portable Internet of Things (IoT)-based EMG system utilizing Message Queuing Telemetry Transport (MQTT) protocol for wireless muscle activity monitoring. The system integrates an EMG V3 sensor module with NodeMCU ESP8266 microcontroller, powered by dual 18650 batteries. Implementation utilized Arduino IDE and IoT MQTT Panel application. Validation comprised functional suitability testing (ISO/IEC 25010:2012), signal characteristics analysis across five subjects with 10 trials each, and repeatability precision evaluation (ISO 17025:2017). The system demonstrated 100% functional suitability. EMG signal characteristics showed average peak-to-peak voltage of 8.95±0.62 mV during relaxation and 10.48±0.58 mV during contraction, with Root Mean Square (RMS) voltage of 1.38±0.11 mV and 1.94±0.21 mV, respectively. Signal frequency-maintained consistency at approximately 50 Hz. Overall repeatability precision achieved 97.62%. This portable EMG system addresses conventional system limitations through wireless connectivity and reduced cost while maintaining measurement accuracy suitable for muscle health monitoring applications.
Classification of mammographic image based on texture features with Random Forest method for identification of breast tumors Rizal Krisdiyanto; Heni Sumarti
Journal of Holistic Medical Technologies (JHMT) Vol. 1 No. 1 (2024): December
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Breast cancer is one of the most common sorts of cancer in ladies around the world. One method that can be used to detect breast cancer is to use medical imaging. Purpose: This study was conducted to identify the types of breast tumors by combining feature extraction and classification using the Random Forest method. The research data comes from the DDSM repository, which consists of 20 benign and 20 malignant images aged 40 to 50 years. The research stages consist of preprocessing, feature extraction, and classification. The preprocessing and feature extraction stages use MATLAB R2021a, while the classification process uses the WEKA application. Texture feature extraction methods include Histograms and GLCM (Gray-Level Co-Occurrence Matrix). The histogram feature extraction results show that the benign image has a higher level of brightness and contrast and is symmetrical compared to the malignant image. Meanwhile, the malignant image has a more random or irregular histogram than the benign image. Then, the average value of GLCM texture features resulting from benign images is higher than malignant images. The texture feature-based breast tumor classification process using the Random Forest method from the Training Set stage obtained an accuracy of 100%. Meanwhile, at the cross-validation stage with variations of 5-Folds, 10-Folds, and 15-Folds, the same value was obtained for an accuracy of 95%. This shows that the Random Forest classification method can be used to identify breast tumor types with more accurate results and does not depend on an individual's ability to read medical imaging results.
Classification of diabetic retinopathy and normal fundus images based on texture features using Multilayer Perceptron (MLP) Ayu Wulandari; Heni Sumarti
Journal of Holistic Medical Technologies (JHMT) Vol. 1 No. 2 (2025): June
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Diabetic retinopathy is a disease caused by uncontrolled blood sugar levels and occurs continuously. Funduscopic examination with an ophthalmoscope tool to determine diabetic retinopathy. This study aims to classify funduscopy images in distinguishing normal eyes and diabetic retinopathy based on texture characteristics using the multilayer perceptron (MLP) method. Texture feature extraction as a class recognition process that aims to produce characteristics based on the texture of each image. The texture features used are histogram and GLCM with 10 parameters. Research data is sourced from the Zenodo website and the National Library of Medicine. Based on the results of the study, it shows that the multilayer perceptron method with the help of Weka machine learning in classifying eye fundus images to distinguish normal eye cases and diabetic retinopathy produces an accuracy value of 83.75% at k-folds 20 cross validation with sensitivity and specificity values of 49.20% and 95.09%.
Identification of istihadhah blood color using TCS3200 color sensor with Multilayer Perceptron (MLP) method Syukrotus Sa’diati; Heni Sumarti
Journal of Holistic Medical Technologies (JHMT) Vol. 1 No. 2 (2025): June
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

One of the natures of women, including menstruation, namely the discharge of blood from the vagina periodically and has a certain cycle. However, this cycle can experience disturbances, in Islam it is referred to as istihadloh. Women who are experiencing istihadloh are called mustahadloh. One of the mustahadloh categories is Mu'tadah Ghoiru Mumayyizah, namely women who have recently bled but cannot distinguish between strong and weak blood. This study aims to distinguish between menstrual blood and istihadloh based on its color. The research method was carried out by designing and manufacturing the TCS3200 color sensor tool based on Arduino Uno with RGB value output results. The average RGB value of istihadloh blood is R 71.6, G 83.7 and B 55.2. While the average RGB values for menstrual blood color are R 143.3, G 176.6 and B 79.3. The results of the classification accuracy using the MLP method in differentiating menstrual blood and istihadloh based on color using the TCS3200 color sensor are 96.7%.
Dose Analysis of Boron Neutron Capture Therapy (BNCT) in Brain Cancer Based on Cyclotron Using PHITS Application Simulation Muhammad Ghozali; Heni Sumarti
Journal of Holistic Medical Technologies (JHMT) Vol. 1 No. 2 (2025): June
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

One of the cancers is brain cancer, the most dangerous of which is Glioblastoma Multiforme (GBM). The research conducted aims to determine the effect of boron concentration on the boron dose rate, irradiation time and absorbed dose. Current cancer treatment still provides deterministic effects (tissue reactions) of radiation to patients and long therapy times. Therefore, researchers conducted research on Boron Neutron Capture Therapy (BNCT) in the treatment of cancer patients that is more selective in destroying cancer cells and is safe because it does not damage healthy tissues around it and the therapy requires a short time. The type of research conducted is quantitative experimental research. The research method with simulation uses the Particle and Heavy Ions Transport code System (PHITS) application on therapy process of Boron Neutron Capture Therapy (BNCT) for brain cancer patients type of Glioma grade 4 called Glioblastoma Multiforme. Patient modeling is based on the Oak Ridge National Laboratory-Medical Internal Radiation Dose (ORNL-MIRD) phantom in adult men who have a brain cancer diameter of 4 cm at a depth of 7 cm with a neutron source from a 30 MeV Cyclotron. The boron concentration used has 3 variations, as follows 20 μg/g, 40 μg/g and 60 μg/g of cancer tissue. Based on the results of the study at a boron concentration of 60 μg/g in the Gross Total Volume (GTV) organ or the center of cancer cells with a dose rate value of 11,160 x 10-2 Gy/s, thus accelerating the irradiation time within a period of 4 minutes 48 seconds and the absorbed dose increases by 30 Gy. Thus, it can be concluded that the higher the boron concentration, the faster the boron dose rate value, the greater the absorbed dose and the faster the irradiation time when carrying out brain cancer therapy.
Experimental Study of Brain Activity on ECVT-Based Motor Movements Hani Nur Endah; Heni Sumarti; Marlin Ramadhan Baidillah
Journal of Holistic Medical Technologies (JHMT) Vol. 2 No. 1 (2025): December
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study investigates the characterization of brain signals in response to motor stimulation using the principle of Electrical Capacitance Volume Tomography (ECVT). The experiment was conducted on a 21-year-old male subject, who was exposed to motor tasks (hand gripping and imagined movement) and audiovisual stimulation (watching a film), alongside control conditions with water and air. Data acquisition was performed using an ECVT helmet sensor at frequencies of 500 kHz, 1 MHz, and 5 MHz. The results revealed variations in vpp values across different conditions, with the highest sensitivity observed at 500 kHz during imagined movement. These findings indicate that internal motor imagery elicits stronger brain activity compared to resting or external stimulation. Overall, the study demonstrates that ECVT is capable of distinguishing brain signal characteristics under different types of stimulation, with 500 kHz identified as the optimal frequency for sensitivity.
Study of the Impact of Dzun Nuun Prayer on Brain Wave Patterns of Adolescents with Overthinking Symptoms Through Electroencephalography (EEG) Examination Putri Zulfikah; Heni Sumarti; Istikomah
Journal of Holistic Medical Technologies (JHMT) Vol. 2 No. 1 (2025): December
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Overthinking is a psychological disorder characterized by excessive, repetitive thoughts. The Dzun Nuun prayer is believed to be a spiritual approach capable of calming the mind and can be examined through the activity of alpha and beta brain waves. This study aims to determine the normalized intensities of alpha and beta waves before and after the administration of the Dzun Nuun prayer stimulus, and to assess its effect on overthinking among adolescents. The research employed a quantitative, experimental method using an Electroencephalography (EEG) device. The subjects were 20 final-year university students experiencing overthinking, and the data were analyzed using a Paired-Samples T-Test. The results showed increased intensities of alpha waves (0.05–0.30) and beta waves (0.01–0.13) following the Dzun Nuun prayer stimulus. Statistical testing revealed significant effects on both alpha (p = 0.026) and beta waves (p = 0.039), indicating a meaningful influence. In conclusion, the Dzun Nuun prayer has a positive effect on brain activity, reducing overthinking symptoms by enhancing relaxation and focus-related brain waves.
Examination of uric acid levels in Karonsih, Semarang: invasive and non-invasive approaches based on research Heni Sumarti; Irman Said Prastyo; Rina Susi Cahyawati; Istikomah; Ice Uliya Sari; Mohammad Candra Malindo
Community Research and Application Journal (CRAJ) Vol. 1 No. 1 (2024): December
Publisher : Konsorsium Pengetahuan Innoscientia (KOPINNOS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Regularly monitoring uric acid levels is essential for preventing and managing gout, where excess uric acid accumulates in joints and causes pain and swelling. Traditionally, invasive blood sample methods are used for this measurement, which can be uncomfortable and increase medical waste. This community service project, organized by the Physics Department of Universitas Islam Negeri Walisongo Semarang in collaboration with the Ngabdi Neliteni Ngabekti (N3) community organization, compares invasive and non-invasive approaches for uric acid level assessment among residents in Karonsih, Semarang. A newly developed non-invasive device was evaluated alongside standard invasive methods based on participant comfort, accuracy, and feedback. Results indicate that, although the non-invasive device showed a mean error rate of 40.52%, it was preferred for its comfort and non-invasive nature. These findings highlight the potential of non-invasive technology in community health monitoring and underline the importance of regular health check-ups facilitated by collaborations between community organizations, educational institutions, and government bodies.