MyMedR

Displaying all 4 publications

Abstract:

Sort:

State-of-the-art ensemble learning and unsupervised learning in fatigue crack recognition of glass fiber reinforced polyester composite (GFRP) using acoustic emission

Gholizadeh S, Leman Z, Baharudin BTHT

Ultrasonics, 2023 Jul;132:106998.
PMID: 37001339 DOI: 10.1016/j.ultras.2023.106998

Fatigue strength is one of the most important properties of composite materials because it directly relates to their lifespan. Acoustic emission (AE) is a passive structural health monitoring (SHM) technique that provides real-time damage detection based on stress waves generated by cracks in the structure. This study evaluates the damage progression on glass fiber reinforced polyester composite specimens using different approaches of machine learning. Different methodologies for damage detection and characterization of AE parameters are presented. Three different ensemble learning methods namely, XGboost, LightGBM, and CatBoost were chosen to predict damages and AE parameters. SHAP values were used to select AE key features and K-means algorithms were employed to classify damage severity. The accuracy of these approaches demonstrates the reliability of various machine learning techniques in predicting the fatigue life of composite materials using acoustic emission.
Simulation study on quality of CPR between manual chest compression and mechanical chest compression devices performed in ambulance

Kamauzaman THT, Ngu JTH, Arithra A, Noh AYM, Siti-Azrin AH, Nor J

Med J Malaysia, 2021 03;76(2):171-176.
PMID: 33742624

BACKGROUND: Maintaining good quality CPR while transporting out-of-hospital cardiac arrest patients is very challenging. We aim to determine how different ambulance speed can affect the quality of chest compression performed either manually or mechanically.
METHODS: This was an observational manikin-based study. A total of 96 participants as well as two types of mechanical compression devices: Lucas-2 and AutoPulse, performed one minute of continuous chest compression on BT-CPEA programmed manikin while the ambulance travelled at different speeds, i.e., idle state, 30km/hr and 60km/hr. Seven outcome variables of chest compression were measured. Performance data of different groups of compressor were compared and analysed using repeated measures analysis of variance (ANOVA).
RESULTS: In manual chest compression, significant variation were noted among different speeds in term of average compression rate (p<0.001), average compression depth (p=0.007), fraction of adequate/insufficient compression depth and fraction of normal hands positioning with p=0.018, 0.022 and 0.034 respectively. Overall, AutoPulse and Lucas-2 were not affected by ambulance speed. Lucas- 2 showed more consistent average compression rate, higher fraction of adequate compression depth and reduced fraction of insufficient compression depth as compared to manual compression with p<0.001, 0.001 and 0.043 respectively.
CONCLUSION: In this study we found that ambulance speed significantly affected certain aspects of manual chest compression most notably compression depth, rate and hand positioning. AutoPulse and Lucas-2 can improve these aspects by providing more consistent compression rate, depth and fraction of adequate compression depth during transport.
The utility of a 'non-significant' coronary angiogram

Wang TK, Oh TH, Samaranayake CB, Webster MW, Stewart JT, Watson T, et al.

Int J Clin Pract, 2015 Dec;69(12):1465-72.
PMID: 26304046 DOI: 10.1111/ijcp.12723

Coronary angiography is the gold standard for assessing coronary artery disease (CAD). In many patients with chest pain, no or mild CAD (< 50% stenosis) is found. It is uncertain whether this 'non-significant' result influences management and outcomes. We reviewed characteristics and outcomes in a contemporary cohort of chest pain referrals who had mild or absent CAD on coronary angiography.
NAIP-deletion analysis in Malaysian patients with spinal muscular atrophy

Watihayati MS, M S W, Zabidi AM, A M H ZH, Tang TH, T H T, et al.

Kobe J Med Sci, 2007;53(4):171-5.
PMID: 17932457

Spinal Muscular Atrophy (SMA) is an autosomal recessive disease, which is characterized by degeneration of the anterior horn cells of the spinal cord. SMA is classified into 3 clinical subtypes, type I (severe), type II (intermediate), and type III (mild). Two genes, SMN1 and NAIP, have been identified as SMA-related genes. The SMN1 gene is now recognized as a responsible gene for the disease because it is deleted or mutated in most SMA patients. However, the role of the NAIP gene in SMA has not been fully clarified. To clarify the contribution of NAIP to the disease severity of SMA, we studied the relationship between NAIP-deletion and clinical phenotype in Malaysian patients. A total of 39 patients lacking SMN1 (12 type I, 19 type II, and 8 type III patients) were enrolled into this study. Seven out of 12 patients with type I SMA (approximately 60%) showed NAIP deletion. On the contrary, only 2 out of 20 type II patients and none of type III patients showed NAIP deletion. There was a statistically significant difference in NAIP-deletion frequency among the clinical subtypes (Fisher's exact probability test, p value = 0.014). In conclusion, according to our data that NAIP deletion was more frequent in type I SMA than in type II-III SMA, the NAIP gene may be a modifying factor for disease severity of SMA.