Asian Journal Of Medical Technology

AUTOMATED DETECTION AND CLASSIFICATION OF BRAIN STROKE LESIONS IN MRI USING MACHINE LEARNING TECHNIQUES

ADAM SAMSUDIN, Norhashimah Mohd Saad, Abdul Rahim Abdullah, Ahmad Sobri Muda — Sat, 31 May 2025 00:00:00 +0000

Magnetic Resonance Imaging (MRI) is crucial for diagnosing brain disorders, with strokes being particularly significant. Recent studies emphasize the importance of prompt stroke treatment, encapsulated in the adage "time is brain," which highlights that intervention within the first six hours can greatly improve outcomes and save lives. However, traditional manual stroke diagnosis by neuroradiologists is often subjective and time-consuming. To address this, our study presents an automated method for detecting, segmenting, and classifying brain stroke lesions from MRI scans. Utilizing machine learning, particularly diffusion-weighted imaging (DWI) sequences, our approach involves four main stages: pre-processing, segmentation, feature extraction, and classification. We employ k-Means for segmentation to identify stroke regions, and statistical features derived from these segments are used for classification with linear discriminant analysis (LDA), support vector machine (SVM), weighted k-Nearest Neighbor (k-NN), and a bagged tree classifier. Performance metrics include the Jaccard index, Dice coefficient, false positive and negative rates for segmentation, and accuracy, sensitivity, and specificity for classification. Our findings show that k-Means is optimal for stroke lesion segmentation, achieving a Dice index of 0.85, while SVM demonstrates the highest classification accuracy of 98.5% with an average training time of 1.8 seconds, suggesting an efficient automated solution for timely and accurate stroke diagnosis.

FABRICATION OF GERMANIUM-DOPED SILICA OPTICAL FIBRES USING MCVD TECHNIQUES TO ENHANCE DOSIMETRIC CHARACTERISTICS FOR GAMMA-RAY DETECTION IN BLOOD IRRADIATOR

Sat, 31 May 2025 00:00:00 +0000

This study explores the fabrication of germanium-doped silica optical fibers using Modified Chemical Vapor Deposition (MCVD) techniques to enhance dosimetric properties for gamma-ray detection in blood irradiators. Accurate dose mapping, essential for preventing graft-versus-host disease, requires radiation exposure in the range of 15 Gy to 50 Gy. We investigate the dosimetric characteristics of Ge-doped silica fibers as passive radiation dosimeters for blood irradiators using a Cs-137 source.

Key dosimetric parameters, including linearity, sensitivity, fading, and glow curve analysis, were tested on cylindrical (CF) and flat (FF) fibers, irradiated with doses from 5 to 50 Gy. Thermoluminescence (TL) measurements showed a linear dose response (R² > 99%) and consistent sensitivity, with Coefficients of Variation improving from 12.8% to 5.57% for CF and 13.8% to 5.91% for FF. After 28 days of storage, TL fading was 11.5% for CF and 9.2% for FF.

The MCVD technique, along with germanium doping, significantly improved the dosimetric properties of the fibers, making both CF and FF Ge-doped optical fibers promising candidates for passive dosimetry in gamma-ray blood irradiators.

ASSESSMENT OF NEPHELIUM LAPPACEUM PEEL EXTRACT AS A POTENTIAL PRIMARY STAIN FOR GRAM STAINING IN BACTERIAL IDENTIFICATION

Sat, 31 May 2025 00:00:00 +0000

The study determines the potential of Nephelium lappaceum peel extract as an alternative to the standard Crystal violet in bacterial identification. This study utilized a quasi-experimental design to assess the staining efficacy of Nephelium lappaceum peel extract. Conventional ethanolic extraction is the method for obtaining the rambutan peel extract. The extract was then turned into a powdered form and portioned to produce concentrations of 25%, 50%, and 75%. The staining process employed was the same as the routine gram staining procedure. The stained specimens were then examined under a light microscope and photos were taken. The photos were subjected to RGB analysis using Adobe Photoshop software. Findings revealed that all dependent variables exhibit significant differences among the groups, as indicated by the provided F-values and significance levels (p = 0.0001425, below the 0.05 threshold)The 75% concentration of Nephelium lappaceum peel extract was the most effective, producing a color closely resembling standard Crystal Violet. This suggests its potential as an alternative primary stain in Gram staining. However, the rambutan peel stain did not effectively stain all Gram-positive bacteria across all tested concentrations. Therefore, to increase the yield of anthocyanins, it is recommended to explore non-conventional extraction methods.

DIAGNOSTIC VALUES OF SONOGRAPHIC ELASTOGRAPHY PARAMETERS IN BREAST LESIONS

Nguyen Vu Dang — Sat, 31 May 2025 00:00:00 +0000

Background: Ultrasound is one of the imaging techniques used to screen and evaluate breast lesions. In particular, when combined with strain elastography, it has been shown to improve the diagnostic effectiveness of breast lesions in many studies. However, few studies have directly compared all three strain elastography methods to determine whether one is more accurate than the others.

Objectives: We aimed to evaluate and compare the sensitivity, specificity, and diagnostic accuracy of the five-point elastography score (Tsukuba score), strain ratio (SR), and length ratio (E/B ratio) in suspected breast lesions correlate with pathology results as the gold standards.

Materials and methods: This is a cross-sectional descriptive study, 59 patients with 61 lesions were classified BIRADS from 3 to 5 after obtaining Strain elastography and B-mode ultrasound that have got histopathologic results at Can Tho Oncology Hospital from November 2022 to May 2024.

Results: In this study, there was a total of 61 lesions of 59 female patients aged 47.2±1.9 years, 26 were benign, and 35 were malignant, they mainly located in the upper outer quadrant. Sensitivity, specificity and accuracy rate of 86%, 81%, and 84% for the Tsukuba score. These of 97%, 79%, and 92% for E/B ratio (cutoff value 1.05, AUC 0.91), 60%, 89%, and 72% for SR (cutoff value 3.09, AUC 0.74) were obtained, respectively.

Conclusion: This study has shown that all three methods are valuable in evaluating breast lesions. Among them, the E/B ratio had the highest sensitivity and diagnostic accuracy, while the specificity of the strain ratio was superior to the others.

A review of research on the relationship between tea drinking and breast cancer

jingjing yu — Sat, 31 May 2025 00:00:00 +0000

The incidence of breast cancer remains high worldwide and is increasing each year in some regions. Despite the variety of current treatments for breast cancer, the preventive and therapeutic effects are still limited due to the highly heterogeneous nature and complex biological mechanisms of breast cancer. In recent years, tea consumption has emerged as a research focus because of its possible anti-cancer properties. Numerous preclinical studies have demonstrated that regular tea intake could potentially curb the progression of breast cancer by influencing various biological mechanisms, including signaling pathways, cell cycle regulation, and immune system responses, among others. Nonetheless, the findings from epidemiological studies show considerable variability, and the connection between tea drinking and both the risk and outlook for breast cancer is shaped by numerous elements. These include the specific type of tea consumed, the quantity consumed, individual genetic variations, and environmental influences.This article sums up the current research findings. It delves into the connection between tea consumption and the risk as well as prognosis of breast cancer among different regional populations. Meanwhile, it expounds on the potential molecular biological mechanisms behind it. The aim is to offer a theoretical foundation for the personalized prevention and treatment of breast cancer down the road.