MyMedR

Displaying publications 41 - 60 of 108 in total

Abstract:

Sort:

Fulltext Machine-Learning-Based Diabetes Mellitus Risk Prediction Using Multi-Layer Neural Network No-Prop Algorithm

Sonia JJ, Jayachandran P, Md AQ, Mohan S, Sivaraman AK, Tee KF

Diagnostics (Basel), 2023 Feb 14;13(4).
PMID: 36832207 DOI: 10.3390/diagnostics13040723

Over the past few decades, the prevalence of chronic illnesses in humans associated with high blood sugar has dramatically increased. Such a disease is referred to medically as diabetes mellitus. Diabetes mellitus can be categorized into three types, namely types 1, 2, and 3. When beta cells do not secrete enough insulin, type 1 diabetes develops. When beta cells create insulin, but the body is unable to use it, type 2 diabetes results. The last category is called gestational diabetes or type 3. This happens during the trimesters of pregnancy in women. Gestational diabetes, however, disappears automatically after childbirth or may continue to develop into type 2 diabetes. To improve their treatment strategies and facilitate healthcare, an automated information system to diagnose diabetes mellitus is required. In this context, this paper presents a novel system of classification of the three types of diabetes mellitus using a multi-layer neural network no-prop algorithm. The algorithm uses two major phases in the information system: the training phase and the testing phase. In each phase, the relevant attributes are identified using the attribute-selection process, and the neural network is trained individually in a multi-layer manner, starting with normal and type 1 diabetes, then normal and type 2 diabetes, and finally healthy and gestational diabetes. Classification is made more effective by the architecture of the multi-layer neural network. To provide experimental analysis and performances of diabetes diagnoses in terms of sensitivity, specificity, and accuracy, a confusion matrix is developed. The maximum specificity and sensitivity values of 0.95 and 0.97 are attained by this suggested multi-layer neural network. With an accuracy score of 97% for the categorization of diabetes mellitus, this proposed model outperforms other models, demonstrating that it is a workable and efficient approach.
Fulltext Accuracy Analysis of Deep Learning Methods in Breast Cancer Classification: A Structured Review

Yusoff M, Haryanto T, Suhartanto H, Mustafa WA, Zain JM, Kusmardi K

Diagnostics (Basel), 2023 Feb 11;13(4).
PMID: 36832171 DOI: 10.3390/diagnostics13040683

Breast cancer is diagnosed using histopathological imaging. This task is extremely time-consuming due to high image complexity and volume. However, it is important to facilitate the early detection of breast cancer for medical intervention. Deep learning (DL) has become popular in medical imaging solutions and has demonstrated various levels of performance in diagnosing cancerous images. Nonetheless, achieving high precision while minimizing overfitting remains a significant challenge for classification solutions. The handling of imbalanced data and incorrect labeling is a further concern. Additional methods, such as pre-processing, ensemble, and normalization techniques, have been established to enhance image characteristics. These methods could influence classification solutions and be used to overcome overfitting and data balancing issues. Hence, developing a more sophisticated DL variant could improve classification accuracy while reducing overfitting. Technological advancements in DL have fueled automated breast cancer diagnosis growth in recent years. This paper reviewed studies on the capability of DL to classify histopathological breast cancer images, as the objective of this study was to systematically review and analyze current research on the classification of histopathological images. Additionally, literature from the Scopus and Web of Science (WOS) indexes was reviewed. This study assessed recent approaches for histopathological breast cancer image classification in DL applications for papers published up until November 2022. The findings of this study suggest that DL methods, especially convolution neural networks and their hybrids, are the most cutting-edge approaches currently in use. To find a new technique, it is necessary first to survey the landscape of existing DL approaches and their hybrid methods to conduct comparisons and case studies.
Fulltext Potential Value of Visfatin, Omentin-1, Nesfatin-1 and Apelin in Renal Cell Carcinoma (RCC): A Systematic Review and Meta-Analysis

Chinapayan SM, Kuppusamy S, Yap NY, Perumal K, Gobe G, Rajandram R

Diagnostics (Basel), 2022 Dec 06;12(12).
PMID: 36553076 DOI: 10.3390/diagnostics12123069

Renal cell carcinoma (RCC) is the most lethal genitourinary malignancy. Obesity is a risk factor for RCC development. The role of adipokines in the relationship between obesity and RCC requires confirmatory evidence in the form of a systematic review and meta-analysis, specifically for visfatin, omentin-1, nesfatin-1 and apelin. A search of databases up to July 2022 (PubMed, Web of Science and Scopus) for studies reporting the association of these selected adipokines with RCC was conducted. A total of 13 studies fulfilled the selection criteria. Only visfatin (p < 0.05) and nesfatin-1 (p < 0.05) had a significant association with RCC. Meanwhile, apelin and omentin-1 showed no association with RCC. The meta-analysis results of nesfatin-1 showed no association with early-stage (OR = 0.09, 95% CI = −0.12−0.29, p = 0.41), late-stage (OR = 0.36, 95% CI = 0.07−1.89, p = 0.23) and low-grade (OR = 1.75, 95% CI = 0.37−8.27, p = 0.48) RCC. However, nesfatin-1 showed an association with a high grade of the disease (OR = 0.29, 95% CI = 0.13−0.61, p = 0.001) and poorer overall survival (OS) (HR = 3.86, 95% CI = 2.18−6.85; p < 0.01). Apelin showed no association with the risk of RCC development (mean difference = 21.15, 95% CI = −23.69−65.99, p = 0.36) and OS (HR = 1.04, 95% Cl = 0.45−2.41; p = 0.92). Although the number of studies evaluated was limited, analysis from this systematic review and meta-analysis indicate that visfatin and nesfatin-1 were elevated. In summary, these adipokines may play a role in the development and progression of RCC and hence may have potential diagnostic and prognostic capabilities for RCC.
Fulltext High EZH2 Protein Expression Is a Poor Prognostic Predictor in IDH1 R132H-Negative Gliomas

Wong YP, Che Abdul Aziz R, Noor Aizuddin A, Mohd Saleh MF, Mohd Arshad R, Tan GC

Diagnostics (Basel), 2022 Sep 30;12(10).
PMID: 36292072 DOI: 10.3390/diagnostics12102383

Accumulating data indicates that enhancer of zeste homology 2 (EZH2) and isocitrate dehydrogenase 1 (IDH1) are implicated in promoting tumourigenesis in a myriad of malignancies including gliomas. We aimed to determine the immunoexpression of EZH2 in gliomas and its correlation with clinicopathological variables. The prognostic value of the combined immunoexpression of EZH2 and IDH1 was further explored in a retrospective analysis involving 56 patients with histologically confirmed gliomas in Universiti Kebangsaan Malaysia Medical Centre from 2010 to 2016. The patients were then followed up for a period of five years. EZH2 and IDH1 R132H immunoexpressions were performed and analysed on respective tissue blocks. Five-year progression-free survival (PFS) and overall survival (OS) were estimated by Kaplan−Meier analysis. Univariate and multivariate Cox proportional hazard regression models were performed to evaluate the value of EZH2 as an independent factor for the prediction of PFS and OS. High EZH2 immunoexpression was demonstrated in 27 (48.2%) gliomas. High EZH2 expression was significantly correlated with older age (p = 0.003), higher tumour grade (p < 0.001), negative IDH1 R132H immunoexpression (p = 0.039), a poor 5-year PFS (mean = 9.7 months, p < 0.001) and 5-year OS (mean = 28.2 months, p = 0.007). In IDH1 R132H-negative gliomas, there was a trend toward shorter 5-year PFS (mean = 8.0 months, p = 0.001) and 5-year OS (mean = 28.7 months, p = 0.06) in gliomas demonstrating high EZH2 expression compared with those with low EZH2 expression. High EZH2 immunoexpression is an unfavourable independent prognostic predictor of poor survival in gliomas. EZH2 analysis might therefore be of clinical value for risk stratification, especially in patients with IDH1 R132H-negative gliomas.
Loss of CXC-Chemokine Receptor 1 Expression in Chorioamnionitis Is Associated with Adverse Perinatal Outcomes

Wong YP, Wagiman N, Tan JW, Hanim BS, Rashidan MSH, Fong KM, et al.

Diagnostics (Basel), 2022 Apr 01;12(4).
PMID: 35453930 DOI: 10.3390/diagnostics12040882

Background: Chorioamnionitis complicates about 1−5% of deliveries at term and causes about one-third of stillbirths. CXC-chemokine receptor 1 (CXCR1) binds IL-8 with high affinity and regulates neutrophil recruitment. We aimed to determine the immunoexpression of CXCR1 in placentas with chorioamnionitis, and its association with adverse perinatal outcomes. Methods: A total of 101 cases of chorioamnionitis and 32 cases of non-chorioamnionitis were recruited over a period of 2 years. CXCR1 immunohistochemistry was performed, and its immunoexpression in placentas was evaluated. The adverse perinatal outcomes included intrauterine death, poor APGAR score, early neonatal death, and respiratory complications. Results: Seventeen cases (17/101, 16.8%) with chorioamnionitis presented as preterm deliveries. Lung complications were more common in mothers who were >35 years (p = 0.003) and with a higher stage in the foetal inflammatory response (p = 0.03). Notably, 24 cases (23.8%) of histological chorioamnionitis were not detected clinically. Interestingly, the loss of CXCR1 immunoexpression in the umbilical cord endothelial cells (UCECs) was significantly associated with foetal death (p = 0.009). Conclusion: The loss of CXCR1 expression in UCECs was significantly associated with an increased risk of adverse perinatal outcomes and could be used as a biomarker to predict adverse perinatal outcomes in chorioamnionitis. Further study is warranted to study the pathophysiology involved in the failure of CXCR1 expression in these cells.
Fulltext Reproducibility and Repeatability of Coronary Computed Tomography Angiography (CCTA) Image Segmentation in Detecting Atherosclerosis: A Radiomics Study

Yunus MM, Sabarudin A, Karim MKA, Nohuddin PNE, Zainal IA, Shamsul MSM, et al.

Diagnostics (Basel), 2022 Aug 19;12(8).
PMID: 36010355 DOI: 10.3390/diagnostics12082007

Atherosclerosis is known as the leading factor in heart disease with the highest mortality rate among the Malaysian population. Usually, the gold standard for diagnosing atherosclerosis is by using the coronary computed tomography angiography (CCTA) technique to look for plaque within the coronary artery. However, qualitative diagnosis for noncalcified atherosclerosis is vulnerable to false-positive diagnoses, as well as inconsistent reporting between observers. In this study, we assess the reproducibility and repeatability of segmenting atherosclerotic lesions manually and semiautomatically in CCTA images to identify the most appropriate CCTA image segmentation method for radiomics analysis to quantitatively extract the atherosclerotic lesion. Thirty (30) CCTA images were taken retrospectively from the radiology image database of Hospital Canselor Tuanku Muhriz (HCTM), Kuala Lumpur, Malaysia. We extract 11,700 radiomics features which include the first-order, second-order and shape features from 180 times of image segmentation. The interest vessels were segmentized manually and semiautomatically using LIFEx (Version 7.0.15, Institut Curie, Orsay, France) software by two independent radiology experts, focusing on three main coronary blood vessels. As a result, manual segmentation with a soft-tissuewindowing setting yielded higher repeatability as compared to semiautomatic segmentation with a significant intraclass correlation coefficient (intra-CC) 0.961 for thefirst-order and shape features; intra-CC of 0.924 for thesecond-order features with p < 0.001. Meanwhile, the semiautomatic segmentation has higher reproducibility as compared to manual segmentation with significant interclass correlation coefficient (inter-CC) of 0.920 (first-order features) and a good interclass correlation coefficient of 0.839 for the second-order features with p < 0.001. The first-order, shape order and second-order features for both manual and semiautomatic segmentation have an excellent percentage of reproducibility and repeatability (intra-CC > 0.9). In conclusion, semi-automated segmentation is recommended for inter-observer study while manual segmentation with soft tissue-windowing can be used for single observer study.
Fulltext An Evaluation of the Quality of Plain Radiograph Interpretations by Radiology Trainees: A Single Institution Experience

A Halim NI, Mohd Zaki F, Manan HA, Mohamed Z

Diagnostics (Basel), 2022 Aug 12;12(8).
PMID: 36010304 DOI: 10.3390/diagnostics12081954

Introduction: The primary communication between the radiologist and referrer is through the radiological report. However, there are incidents of misinterpretation during radiologist training. Therefore, the present study aimed to evaluate the accuracy level and incidence of interpretation errors for plain radiographs among radiology trainees at our institution. Materials and Methods: The present study retrospectively reviewed 508 reported plain radiographs for one year, and two radiologists subsequently evaluated these plain radiographs. The initial diagnosis by the trainee was compared with the radiologists’ evaluation, and the results were categorized as either ‘accurate’, ‘minor discrepancy’, or ‘major discrepancy’. The data were analyzed concerning the overall performance, year of trainee, anatomic area, patient age group, and radiograph type. A chi-square test was performed, with p < 0.05 indicating statistical significance. Results: The overall accuracy rate was 69%, with minor and major discrepancy rates of 21% and 10%, respectively. There was an insignificant increase in overall accuracy with increased years of training, despite a reduction to 58% accuracy among Year 3 trainees. The accuracy level increased between Year 1, Year 2 and Year 4 by 70%, 71% and 75%, respectively (p > 0.05). The accuracy rates for both the adult and pediatric age groups were not statistically significant. The mobile radiographs showed lower accuracy rate of reporting than the plain radiographs. Conclusion: The radiological trainee interpretations for plain radiographs had an average rating with low discrepancy rates. The Year 3 trainees had the lowest accuracy compared to the other trainee groups. However, the present study suggests the need for further research to determine if the current outcomes are outliers or are indicative of a real phenomenon.
Fulltext Stability and Reproducibility of Radiomic Features Based on Various Segmentation Techniques on Cervical Cancer DWI-MRI

Ramli Z, Karim MKA, Effendy N, Abd Rahman MA, Kechik MMA, Ibahim MJ, et al.

Diagnostics (Basel), 2022 Dec 12;12(12).
PMID: 36553132 DOI: 10.3390/diagnostics12123125

Cervical cancer is the most common cancer and ranked as 4th in morbidity and mortality among Malaysian women. Currently, Magnetic Resonance Imaging (MRI) is considered as the gold standard imaging modality for tumours with a stage higher than IB2, due to its superiority in diagnostic assessment of tumour infiltration with excellent soft-tissue contrast. In this research, the robustness of semi-automatic segmentation has been evaluated using a flood-fill algorithm for quantitative feature extraction, using 30 diffusion weighted MRI images (DWI-MRI) of cervical cancer patients. The relevant features were extracted from DWI-MRI segmented images of cervical cancer. First order statistics, shape features, and textural features were extracted and analysed. The intra-class relation coefficient (ICC) was used to compare 662 radiomic features extracted from manual and semi-automatic segmentations. Notably, the features extracted from the semi-automatic segmentation and flood filling algorithm (average ICC = 0.952 0.009, p > 0.05) were significantly higher than the manual extracted features (average ICC = 0.897 0.011, p > 0.05). Henceforth, we demonstrate that the semi-automatic segmentation is slightly expanded to manual segmentation as it produces more robust and reproducible radiomic features.
Fulltext KRAS Mutational Profiles among Colorectal Cancer Patients in the East Coast of Peninsular Malaysia

Hasbullah HH, Sulong S, Che Jalil NA, Abdul Aziz AA, Musa N, Musa M

Diagnostics (Basel), 2023 Feb 21;13(5).
PMID: 36899966 DOI: 10.3390/diagnostics13050822

BACKGROUND: KRAS is a key driver gene in colorectal carcinogenesis. Despite this, there are still limited data on the mutational status of KRAS amongst colorectal cancer (CRC) patients in Malaysia. In the present study, we aimed to analyze the KRAS mutational profiles on codons 12 and 13 amongst CRC patients in Hospital Universiti Sains Malaysia, Kelantan, located on the East Coast of Peninsular Malaysia.
METHODS: DNA were extracted from formalin-fixed, paraffin-embedded tissues obtained from 33 CRC patients diagnosed between 2018 and 2019. Amplifications of codons 12 and 13 of KRAS were conducted using conventional polymerase chain reaction (PCR) followed by Sanger sequencing.
RESULTS: Mutations were identified in 36.4% (12/33) of patients, with G12D (50%) being the most frequent single-point mutation observed, followed by G12V (25%), G13D (16.7%), and G12S (8.3%). No correlation was found between mutant KRAS and location of the tumor, staging, and initial carcinoembryonic antigen (CEA) level.
CONCLUSION: Current analyses revealed that a significant proportion of CRC patients in the East Coast of Peninsular Malaysia have KRAS mutations, where this frequency is higher compared to those in the West Coast. The findings of this study would serve as a precursor for further research that explores KRAS mutational status and the profiling of other candidate genes among Malaysian CRC patients.
Fulltext Gene Mutation Spectrum among Alpha-Thalassaemia Patients in Northeast Peninsular Malaysia

Vijian D, Wan Ab Rahman WS, Ponnuraj KT, Zulkafli Z, Bahar R, Yasin N, et al.

Diagnostics (Basel), 2023 Feb 27;13(5).
PMID: 36900038 DOI: 10.3390/diagnostics13050894

(1) Background: Alpha (α)-thalassaemia is a genetic disorder that affects 5% of the world population. Deletional or nondeletional mutations of one or both HBA1 and HBA2 on chromosome 16 will result in reduced production of α-globin chains, a component of haemoglobin (Hb) that is required for the formation of red blood cells (RBCs). This study aimed to determine the prevalence, haematological and molecular characterisations of α-thalassaemia. (2) Method: The parameters were based on full blood count, high-performance liquid chromatography and capillary electrophoresis. The molecular analysis involved gap-polymerase chain reaction (PCR), multiplex amplification refractory mutation system-PCR, multiplex ligation-dependent probe amplification and Sanger sequencing. (3) Results: With a total cohort of 131 patients, the prevalence of α-thalassaemia was 48.9%, leaving the remaining 51.1% with potentially undetected α gene mutations. The following genotypes were detected: -α3.7/αα (15.4%), -α4.2/αα (3.7%), --SEA/αα (7.4%), αCSα/αα (10.3%), αAdanaα/αα (0.7%), αQuong Szeα/αα (1.5%), -α3.7/-α3.7 (0.7%), αCSα/αCSα (0.7%), -α4.2/αCSα (0.7%), -SEA/αCSα (1.5%), -SEA/αQuong Szeα (0.7%), -α3.7/αAdanaα (0.7%), --SEA/-α3.7 (2.2%) and αCSα/αAdanaα (0.7%). Indicators such as Hb (p = 0.022), mean corpuscular volume (p = 0.009), mean corpuscular haemoglobin (p = 0.017), RBC (p = 0.038) and haematocrit (p = 0.058) showed significant changes among patients with deletional mutations, but not between patients with nondeletional mutations. (4) Conclusions: A wide range of haematological parameters was observed among patients, including those with the same genotype. Thus, a combination of molecular technologies and haematological parameters is necessary for the accurate detection of α-globin chain mutations.
Fulltext Red Blood Cell Alloimmunization and Its Associated Factors among Chronic Liver Disease Patients in a Teaching Hospital in Northeastern Malaysia

S Abdullah SZ, Hassan MN, Ramli M, Abdullah M, Mohd Noor NH

Diagnostics (Basel), 2023 Feb 25;13(5).
PMID: 36900030 DOI: 10.3390/diagnostics13050886

Red blood cell (RBC) alloimmunization is an important complication of blood transfusion. Variations in the frequency of alloimmunization have been noted among different patient populations. We aimed to determine the prevalence of RBC alloimmunization and associated factors among chronic liver disease (CLD) patients in our center. This is a case-control study involving 441 patients with CLD who were being treated at Hospital Universiti Sains Malaysia and subjected to pre-transfusion testing from April 2012 until April 2022. Clinical and laboratory data were retrieved and statistically analyzed. A total of 441 CLD patients were included in our study, with the majority being elderly, with the mean age of patients 57.9 (SD ± 12.1) years old, male (65.1%) and Malays (92.1%). The most common causes of CLD in our center are viral hepatitis (62.1%) and metabolic liver disease (25.4%). Twenty-four patients were reported to have RBC alloimmunization, resulting in an overall prevalence of 5.4%. Higher rates of alloimmunization were seen in females (7.1%) and patients with autoimmune hepatitis (11.1%). Most patients developed a single alloantibody (83.3%). The most common alloantibody identified belonged to the Rh blood group, anti-E (35.7%) and anti-c (14.3%), followed by the MNS blood group, anti-Mia (17.9%). There was no significant factor association of RBC alloimmunization among CLD patients identified. Our center has a low prevalence of RBC alloimmunization among CLD patients. However, the majority of them developed clinically significant RBC alloantibodies, mostly from the Rh blood group. Therefore, phenotype matching for Rh blood groups should be provided for CLD patients requiring blood transfusions in our center to prevent RBC alloimmunization.
H. pylori Related Atrophic Gastritis Detection Using Enhanced Convolution Neural Network (CNN) Learner

Yacob YM, Alquran H, Mustafa WA, Alsalatie M, Sakim HAM, Lola MS

Diagnostics (Basel), 2023 Jan 17;13(3).
PMID: 36766441 DOI: 10.3390/diagnostics13030336

Atrophic gastritis (AG) is commonly caused by the infection of the Helicobacter pylori (H. pylori) bacteria. If untreated, AG may develop into a chronic condition leading to gastric cancer, which is deemed to be the third primary cause of cancer-related deaths worldwide. Precursory detection of AG is crucial to avoid such cases. This work focuses on H. pylori-associated infection located at the gastric antrum, where the classification is of binary classes of normal versus atrophic gastritis. Existing work developed the Deep Convolution Neural Network (DCNN) of GoogLeNet with 22 layers of the pre-trained model. Another study employed GoogLeNet based on the Inception Module, fast and robust fuzzy C-means (FRFCM), and simple linear iterative clustering (SLIC) superpixel algorithms to identify gastric disease. GoogLeNet with Caffe framework and ResNet-50 are machine learners that detect H. pylori infection. Nonetheless, the accuracy may become abundant as the network depth increases. An upgrade to the current standards method is highly anticipated to avoid untreated and inaccurate diagnoses that may lead to chronic AG. The proposed work incorporates improved techniques revolving within DCNN with pooling as pre-trained models and channel shuffle to assist streams of information across feature channels to ease the training of networks for deeper CNN. In addition, Canonical Correlation Analysis (CCA) feature fusion method and ReliefF feature selection approaches are intended to revamp the combined techniques. CCA models the relationship between the two data sets of significant features generated by pre-trained ShuffleNet. ReliefF reduces and selects essential features from CCA and is classified using the Generalized Additive Model (GAM). It is believed the extended work is justified with a 98.2% testing accuracy reading, thus providing an accurate diagnosis of normal versus atrophic gastritis.
Fulltext Enhancement of Ultrasound B-Mode Image Quality Using Nonlinear Filtered-Multiply-and-Sum Compounding for Improved Carotid Artery Segmentation

Mohamed Moubark A, Nie L, Mohd Zaman MH, Islam MT, Zulkifley MA, Baharuddin MH, et al.

Diagnostics (Basel), 2023 Mar 18;13(6).
PMID: 36980469 DOI: 10.3390/diagnostics13061161

In ultrasound B-mode imaging, the axial resolution (AR) is commonly determined by the duration or bandwidth of an excitation signal. A shorter-duration pulse will produce better resolution compared to a longer one but with compromised penetration depth. Instead of relying on the pulse duration or bandwidth to improve the AR, an alternative method termed filtered multiply and sum (FMAS) has been introduced in our previous work. For spatial-compounding, FMAS uses the autocorrelation technique as used in filtered-delay multiply and sum (FDMAS), instead of conventional averaging. FMAS enables a higher frame rate and less computational complexity than conventional plane-wave compound imaging beamformed with delay and sum (DAS) and FDMAS. Moreover, it can provide an improved contrast ratio and AR. In previous work, no explanation was given on how FMAS was able to improve the AR. Thus, in this work, we discuss in detail the theory behind the proposed FMAS algorithm and how it is able to improve the spatial resolution mainly in the axial direction. Simulations, experimental phantom measurements and in vivo studies were conducted to benchmark the performance of the proposed method. We also demonstrate how the suggested new algorithm may be used in a practical biomedical imaging application. The balloon snake active contour segmentation technique was applied to the ultrasound B-mode image of a common carotid artery produced with FMAS. The suggested method is capable of reducing the number of iterations for the snake to settle on the region-of-interest contour, accelerating the segmentation process.
Fulltext Performance Validation of COVID-19 Self-Conduct Buccal and Nasal Swabs RTK-Antigen Diagnostic Kit

Kalil MNA, Yusof W, Ahmed N, Fauzi MH, Bakar MAA, Sjahid AS, et al.

Diagnostics (Basel), 2021 Nov 30;11(12).
PMID: 34943482 DOI: 10.3390/diagnostics11122245

The antigen rapid diagnostic test (Ag-RDT) is an immunodiagnostic test that detects the presence of viral proteins (antigens) expressed by the COVID-19 virus in a sample from a patient's respiratory tract. This study focused on evaluating the performance of self-conduct buccal and nasal swabs RTK-antigen test compared to nasopharyngeal swab RTK-based COVID-19 diagnostic assays, Panbio™ COVID-19 Ag Rapid Test Device (Nasopharyngeal) (Abbott Rapid Diagnostics Jena GmbH, Jena, Germany) used in hospitals for first-line screening. The sensitivity and specificity of the paired RTK-Ag test in detecting the an-tigen were calculated at 96.4% and 100%, respectively. Fisher exact tests showed the association between nasopharyngeal swabs RTK-Ag assay and buccal-nasal swabs RTK-Ag from ProdetectTM is significant (p-values < 0.001). The result showed that a self-conducted buccal and nasal RTK-antigen rapid test by the patients is comparable to the results obtained from a rapid test device conducted by trained medical personnel using a nasopharyngeal swab.
Overexpression of Connexin 40 in the Vascular Endothelial Cells of Placenta with Acute Chorioamnionitis

Tan JY, Yeoh HXY, Chia WK, Tan JW, Aizuddin AN, Farouk WI, et al.

Diagnostics (Basel), 2024 Apr 12;14(8).
PMID: 38667457 DOI: 10.3390/diagnostics14080811

BACKGROUND: Connexins (Cx) 43 and 40 play a role in leukocytes recruitment in acute inflammation. They are expressed in the endothelial cells. They are also found in the placenta and involved in the placenta development. Acute chorioamnionitis is associated with an increased risk of adverse perinatal outcomes. The aim of this study was to determine the expressions of Cx43 and Cx40 in the placenta of mothers with acute chorioamnionitis, and to correlate their association with the severity of chorioamnionitis and adverse perinatal outcomes.
METHODS: This study comprised a total of 81 cases, consisting of 39 placenta samples of mothers with acute chorioamnionitis and 42 non-acute chorioamnionitis controls. Cx43 and Cx40 immunohistochemistry were performed on all cases and their expressions were evaluated on cytotrophoblasts, syncytiotrophoblasts, chorionic villi endothelial cells, stem villi endothelial cells, maternal endothelial cells and decidua of the placenta.
RESULTS: Primigravida has a significantly higher risk of developing acute chorioamnionitis (p < 0.001). Neonates of mothers with a higher stage of fetal inflammatory response was significantly associated with lung complications (p = 0.041) compared to neonates of mothers with a lower stage. The expression of Cx40 was significantly higher in fetal and maternal vascular endothelial cells in acute chorioamnionitis (p < 0.001 and p = 0.037, respectively) compared to controls. Notably, Cx43 was not expressed in most of the types of cells in the placenta, except for decidua. Both Cx43 and Cx40 expressions did not have correlation with the severity of acute chorioamnionitis and adverse perinatal outcomes.
CONCLUSION: Cx40 was overexpressed in the fetal and maternal vascular endothelial cells in the placenta of mothers with acute chorioamnionitis, and it may have a role in the development of inflammation in placenta.
Fulltext Automated Computer-Aided Detection and Classification of Intracranial Hemorrhage Using Ensemble Deep Learning Techniques

Umapathy S, Murugappan M, Bharathi D, Thakur M

Diagnostics (Basel), 2023 Sep 18;13(18).
PMID: 37761354 DOI: 10.3390/diagnostics13182987

Diagnosing Intracranial Hemorrhage (ICH) at an early stage is difficult since it affects the blood vessels in the brain, often resulting in death. We propose an ensemble of Convolutional Neural Networks (CNNs) combining Squeeze and Excitation-based Residual Networks with the next dimension (SE-ResNeXT) and Long Short-Term Memory (LSTM) Networks in order to address this issue. This research work primarily used data from the Radiological Society of North America (RSNA) brain CT hemorrhage challenge dataset and the CQ500 dataset. Preprocessing and data augmentation are performed using the windowing technique in the proposed work. The ICH is then classified using ensembled CNN techniques after being preprocessed, followed by feature extraction in an automatic manner. ICH is classified into the following five types: epidural, intraventricular, subarachnoid, intra-parenchymal, and subdural. A gradient-weighted Class Activation Mapping method (Grad-CAM) is used for identifying the region of interest in an ICH image. A number of performance measures are used to compare the experimental results with various state-of-the-art algorithms. By achieving 99.79% accuracy with an F-score of 0.97, the proposed model proved its efficacy in detecting ICH compared to other deep learning models. The proposed ensembled model can classify epidural, intraventricular, subarachnoid, intra-parenchymal, and subdural hemorrhages with an accuracy of 99.89%, 99.65%, 98%, 99.75%, and 99.88%. Simulation results indicate that the suggested approach can categorize a variety of intracranial bleeding types. By implementing the ensemble deep learning technique using the SE-ResNeXT and LSTM models, we achieved significant classification accuracy and AUC scores.
Fulltext Diagnosing Melanomas in Dermoscopy Images Using Deep Learning

Alwakid G, Gouda W, Humayun M, Jhanjhi NZ

Diagnostics (Basel), 2023 May 22;13(10).
PMID: 37238299 DOI: 10.3390/diagnostics13101815

When it comes to skin tumors and cancers, melanoma ranks among the most prevalent and deadly. With the advancement of deep learning and computer vision, it is now possible to quickly and accurately determine whether or not a patient has malignancy. This is significant since a prompt identification greatly decreases the likelihood of a fatal outcome. Artificial intelligence has the potential to improve healthcare in many ways, including melanoma diagnosis. In a nutshell, this research employed an Inception-V3 and InceptionResnet-V2 strategy for melanoma recognition. The feature extraction layers that were previously frozen were fine-tuned after the newly added top layers were trained. This study used data from the HAM10000 dataset, which included an unrepresentative sample of seven different forms of skin cancer. To fix the discrepancy, we utilized data augmentation. The proposed models outperformed the results of the previous investigation with an effectiveness of 0.89 for Inception-V3 and 0.91 for InceptionResnet-V2.
Fulltext A Hybrid Technique for Diabetic Retinopathy Detection Based on Ensemble-Optimized CNN and Texture Features

Ishtiaq U, Abdullah ERMF, Ishtiaque Z

Diagnostics (Basel), 2023 May 22;13(10).
PMID: 37238304 DOI: 10.3390/diagnostics13101816

One of the most prevalent chronic conditions that can result in permanent vision loss is diabetic retinopathy (DR). Diabetic retinopathy occurs in five stages: no DR, and mild, moderate, severe, and proliferative DR. The early detection of DR is essential for preventing vision loss in diabetic patients. In this paper, we propose a method for the detection and classification of DR stages to determine whether patients are in any of the non-proliferative stages or in the proliferative stage. The hybrid approach based on image preprocessing and ensemble features is the foundation of the proposed classification method. We created a convolutional neural network (CNN) model from scratch for this study. Combining Local Binary Patterns (LBP) and deep learning features resulted in the creation of the ensemble features vector, which was then optimized using the Binary Dragonfly Algorithm (BDA) and the Sine Cosine Algorithm (SCA). Moreover, this optimized feature vector was fed to the machine learning classifiers. The SVM classifier achieved the highest classification accuracy of 98.85% on a publicly available dataset, i.e., Kaggle EyePACS. Rigorous testing and comparisons with state-of-the-art approaches in the literature indicate the effectiveness of the proposed methodology.
Fulltext An Adaptive Early Stopping Technique for DenseNet169-Based Knee Osteoarthritis Detection Model

Al-Rimy BAS, Saeed F, Al-Sarem M, Albarrak AM, Qasem SN

Diagnostics (Basel), 2023 May 29;13(11).
PMID: 37296755 DOI: 10.3390/diagnostics13111903

Knee osteoarthritis (OA) detection is an important area of research in health informatics that aims to improve the accuracy of diagnosing this debilitating condition. In this paper, we investigate the ability of DenseNet169, a deep convolutional neural network architecture, for knee osteoarthritis detection using X-ray images. We focus on the use of the DenseNet169 architecture and propose an adaptive early stopping technique that utilizes gradual cross-entropy loss estimation. The proposed approach allows for the efficient selection of the optimal number of training epochs, thus preventing overfitting. To achieve the goal of this study, the adaptive early stopping mechanism that observes the validation accuracy as a threshold was designed. Then, the gradual cross-entropy (GCE) loss estimation technique was developed and integrated to the epoch training mechanism. Both adaptive early stopping and GCE were incorporated into the DenseNet169 for the OA detection model. The performance of the model was measured using several metrics including accuracy, precision, and recall. The obtained results were compared with those obtained from the existing works. The comparison shows that the proposed model outperformed the existing solutions in terms of accuracy, precision, recall, and loss performance, which indicates that the adaptive early stopping coupled with GCE improved the ability of DenseNet169 to accurately detect knee OA.
Fulltext Heart Failure Emergency Readmission Prediction Using Stacking Machine Learning Model

Rahman MS, Rahman HR, Prithula J, Chowdhury MEH, Ahmed MU, Kumar J, et al.

Diagnostics (Basel), 2023 Jun 02;13(11).
PMID: 37296800 DOI: 10.3390/diagnostics13111948

Heart failure is a devastating disease that has high mortality rates and a negative impact on quality of life. Heart failure patients often experience emergency readmission after an initial episode, often due to inadequate management. A timely diagnosis and treatment of underlying issues can significantly reduce the risk of emergency readmissions. The purpose of this project was to predict emergency readmissions of discharged heart failure patients using classical machine learning (ML) models based on Electronic Health Record (EHR) data. The dataset used for this study consisted of 166 clinical biomarkers from 2008 patient records. Three feature selection techniques were studied along with 13 classical ML models using five-fold cross-validation. A stacking ML model was trained using the predictions of the three best-performing models for final classification. The stacking ML model provided an accuracy, precision, recall, specificity, F1-score, and area under the curve (AUC) of 89.41%, 90.10%, 89.41%, 87.83%, 89.28%, and 0.881, respectively. This indicates the effectiveness of the proposed model in predicting emergency readmissions. The healthcare providers can intervene pro-actively to reduce emergency hospital readmission risk and improve patient outcomes and decrease healthcare costs using the proposed model.

Filters

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links