MyMedR

Displaying all 7 publications

Abstract:

Sort:

Automatic colonic polyp detection using integration of modified deep residual convolutional neural network and ensemble learning approaches

Liew WS, Tang TB, Lin CH, Lu CK

Comput Methods Programs Biomed, 2021 Jul;206:106114.
PMID: 33984661 DOI: 10.1016/j.cmpb.2021.106114

BACKGROUND AND OBJECTIVE: The increased incidence of colorectal cancer (CRC) and its mortality rate have attracted interest in the use of artificial intelligence (AI) based computer-aided diagnosis (CAD) tools to detect polyps at an early stage. Although these CAD tools have thus far achieved a good accuracy level to detect polyps, they still have room to improve further (e.g. sensitivity). Therefore, a new CAD tool is developed in this study to detect colonic polyps accurately.
METHODS: In this paper, we propose a novel approach to distinguish colonic polyps by integrating several techniques, including a modified deep residual network, principal component analysis and AdaBoost ensemble learning. A powerful deep residual network architecture, ResNet-50, was investigated to reduce the computational time by altering its architecture. To keep the interference to a minimum, median filter, image thresholding, contrast enhancement, and normalisation techniques were exploited on the endoscopic images to train the classification model. Three publicly available datasets, i.e., Kvasir, ETIS-LaribPolypDB, and CVC-ClinicDB, were merged to train the model, which included images with and without polyps.
RESULTS: The proposed approach trained with a combination of three datasets achieved Matthews Correlation Coefficient (MCC) of 0.9819 with accuracy, sensitivity, precision, and specificity of 99.10%, 98.82%, 99.37%, and 99.38%, respectively.
CONCLUSIONS: These results show that our method could repeatedly classify endoscopic images automatically and could be used to effectively develop computer-aided diagnostic tools for early CRC detection.
Fulltext Forensic Age Estimation of Chinese Malaysian Adults by Evaluating Occlusal Tooth Wear Using Modified Kim's Index

Lu CK, Yee MCS, Ravi SB, Pandurangappa R

Int J Dent, 2017;2017:4265753.
PMID: 29129975 DOI: 10.1155/2017/4265753

Background and Objective: Evaluation of dental attrition is an easy and relatively accurate approach to estimating the age of an adult either ante- or postmortem for some specific population. Dental attrition represents a progressive physiological age change that can be measured using variety of indices to aid as an adjunct in forensic age estimation. Some of the previously proposed indices have their own practical limitations. This paper focuses on using modified Kim's criteria to score dental attrition to estimate the age of Chinese Malaysian adults and validate it.
Methodology: Tooth wear was evaluated on 190 dental models of Chinese Malaysian adults (age range: 20-60 years) using modified Kim's index to custom-derive a population specific linear equation. The same equation was validated further on new 60 dental casts.
Results and Conclusion: Regression analysis revealed good correlation between age and teeth wear and lower standard error of estimate. Test of regression on a test sample (n = 30 pairs, age range: 20-60 years) showed insignificant difference between predicted versus the actual age with statistically acceptable mean absolute difference. These data suggest that modified Kim's index can be used effectively in forensic age estimation.
Fulltext fNIRS-based functional connectivity estimation using semi-metric analysis to study decision making by nursing students and registered nurses

Chong JS, Chan YL, Ebenezer EGM, Chen HY, Kiguchi M, Lu CK, et al.

Sci Rep, 2020 12 16;10(1):22041.
PMID: 33328535 DOI: 10.1038/s41598-020-79053-z

This study aims to investigate the generalizability of the semi-metric analysis of the functional connectivity (FC) for functional near-infrared spectroscopy (fNIRS) by applying it to detect the dichotomy in differential FC under affective and neutral emotional states in nursing students and registered nurses during decision making. The proposed method employs wavelet transform coherence to construct FC networks and explores semi-metric analysis to extract network redundancy features, which has not been considered in conventional fNIRS-based FC analyses. The trials of the proposed method were performed on 19 nursing students and 19 registered nurses via a decision-making task under different emotional states induced by affective and neutral emotional stimuli. The cognitive activities were recorded using fNIRS, and the emotional stimuli were adopted from the International Affective Digitized Sound System (IADS). The induction of emotional effects was validated by heart rate variability (HRV) analysis. The experimental results by the proposed method showed significant difference (FDR-adjusted p = 0.004) in the nursing students' cognitive FC network under the two different emotional conditions, and the semi-metric percentage (SMP) of the right prefrontal cortex (PFC) was found to be significantly higher than the left PFC (FDR-adjusted p = 0.036). The benchmark method (a typical weighted graph theory analysis) gave no significant results. In essence, the results support that the semi-metric analysis can be generalized and extended to fNIRS-based functional connectivity estimation.
Pulse waveform analysis on temporal changes in ocular blood flow due to caffeine intake: a comparative study between habitual and non-habitual groups

Ismail A, Bhatti MS, Faye I, Lu CK, Laude A, Tang TB

Graefes Arch Clin Exp Ophthalmol, 2018 Sep;256(9):1711-1721.
PMID: 29876732 DOI: 10.1007/s00417-018-4030-9

PURPOSE: To evaluate and compare the temporal changes in pulse waveform parameters of ocular blood flow (OBF) between non-habitual and habitual groups due to caffeine intake.
METHOD: This study was conducted on 19 healthy subjects (non-habitual 8; habitual 11), non-smoking and between 21 and 30 years of age. Using laser speckle flowgraphy (LSFG), three areas of optical nerve head were analyzed which are vessel, tissue, and overall, each with ten pulse waveform parameters, namely mean blur rate (MBR), fluctuation, skew, blowout score (BOS), blowout time (BOT), rising rate, falling rate, flow acceleration index (FAI), acceleration time index (ATI), and resistive index (RI). Two-way mixed ANOVA was used to determine the difference between every two groups where p
Fulltext Automatic Polyp Segmentation in Colonoscopy Images Using a Modified Deep Convolutional Encoder-Decoder Architecture

Eu CY, Tang TB, Lin CH, Lee LH, Lu CK

Sensors (Basel), 2021 Aug 20;21(16).
PMID: 34451072 DOI: 10.3390/s21165630

Colorectal cancer has become the third most commonly diagnosed form of cancer, and has the second highest fatality rate of cancers worldwide. Currently, optical colonoscopy is the preferred tool of choice for the diagnosis of polyps and to avert colorectal cancer. Colon screening is time-consuming and highly operator dependent. In view of this, a computer-aided diagnosis (CAD) method needs to be developed for the automatic segmentation of polyps in colonoscopy images. This paper proposes a modified SegNet Visual Geometry Group-19 (VGG-19), a form of convolutional neural network, as a CAD method for polyp segmentation. The modifications include skip connections, 5 × 5 convolutional filters, and the concatenation of four dilated convolutions applied in parallel form. The CVC-ClinicDB, CVC-ColonDB, and ETIS-LaribPolypDB databases were used to evaluate the model, and it was found that our proposed polyp segmentation model achieved an accuracy, sensitivity, specificity, precision, mean intersection over union, and dice coefficient of 96.06%, 94.55%, 97.56%, 97.48%, 92.3%, and 95.99%, respectively. These results indicate that our model performs as well as or better than previous schemes in the literature. We believe that this study will offer benefits in terms of the future development of CAD tools for polyp segmentation for colorectal cancer diagnosis and management. In the future, we intend to embed our proposed network into a medical capsule robot for practical usage and try it in a hospital setting with clinicians.
Regional comparison on ciguatoxicity, hemolytic activity, and toxin profile of the dinoflagellate Gambierdiscus from Kiribati and Malaysia

Zhu J, Lee WH, Yip KC, Wu Z, Wu J, Leaw CP, et al.

Sci Total Environ, 2023 May 10;872:162236.
PMID: 36791857 DOI: 10.1016/j.scitotenv.2023.162236

The dinoflagellates Gambierdiscus and Fukuyoa can produce Ciguatoxins (CTXs) and Maitotoxins (MTXs) that lead to ciguatera poisoning (CP). The CP hotspots, however, do not directly relate to the occurrence of the ciguatoxic Gambierdiscus and Fukuyoa. Species-wide investigations often showed no association between CTX level and the molecular identity of the dinoflagellates. In the Pacific region, Kiribati is known as a CP hotspot, while Malaysia has only three CP outbreaks reported thus far. Although ciguatoxic strains of Gambierdiscus were isolated from both Kiribati and Malaysia, no solid evidence on the contribution of ciguatoxic strains to the incidence of CP outbreak was recorded. The present study aims to investigate the regional differences in CP risks through region-specific toxicological assessment of Gambierdiscus and Fukuyoa. A total of 19 strains of Gambierdiscus and a strain of Fukuyoa were analyzed by cytotoxicity assay of the neuro-2a cell line, hemolytic assay of fish erythrocytes, and high-resolution mass spectrometry. Gambierdiscus from both Kiribati and Malaysia showed detectable ciguatoxicity; however, the Kiribati strains were more hemolytic. Putative 44-methylgambierone was identified as part of the contributors to the hemolytic activity, and other unknown hydrophilic toxins produced can be potentially linked to higher CP incidence in Kiribati.
Taxonomic assignment of the benthic toxigenic dinoflagellate Gambierdiscus sp. type 6 as Gambierdiscus balechii (Dinophyceae), including its distribution and ciguatoxicity

Dai X, Mak YL, Lu CK, Mei HH, Wu JJ, Lee WH, et al.

Harmful Algae, 2017 07;67:107-118.
PMID: 28755713 DOI: 10.1016/j.hal.2017.07.002

Recent molecular phylogenetic studies of Gambierdiscus species flagged several new species and genotypes, thus leading to revitalizing its systematics. The inter-relationships of clades revealed by the primary sequence information of nuclear ribosomal genes (rDNA), however, can sometimes be equivocal, and therefore, in this study, the taxonomic status of a ribotype, Gambierdiscus sp. type 6, was evaluated using specimens collected from the original locality, Marakei Island, Republic of Kiribati; and specimens found in Rawa Island, Peninsular Malaysia, were further used for comparison. Morphologically, the ribotype cells resembled G. scabrosus, G. belizeanus, G. balechii, G. cheloniae and G. lapillus in thecal ornamentation, where the thecal surfaces are reticulate-foveated, but differed from G. scabrosus by its hatchet-shaped Plate 2', and G. belizeanus by the asymmetrical Plate 3'. To identify the phylogenetic relationship of this ribotype, a large dataset of the large subunit (LSU) and small subunit (SSU) rDNAs were compiled, and performed comprehensive analyses, using Bayesian-inference, maximum-parsimony, and maximum-likelihood, for the latter two incorporating the sequence-structure information of the SSU rDNA. Both the LSU and SSU rDNA phylogenetic trees displayed an identical topology and supported the hypothesis that the relationship between Gambierdiscus sp. type 6 and G. balechii was monophyletic. As a result, the taxonomic status of Gambierdiscus sp. type 6 was revised, and assigned as Gambierdiscus balechii. Toxicity analysis using neuroblastoma N2A assay confirmed that the Central Pacific strains were toxic, ranging from 1.1 to 19.9 fg P-CTX-1 eq cell-1, but no toxicity was detected in a Western Pacific strain. This suggested that the species might be one of the species contributing to the high incidence rate of ciguatera fish poisoning in Marakei Island.