Browse publications by year: 2024

  1. Praveen SP, Hasan MK, Abdullah SNHS, Sirisha U, Tirumanadham NSKMK, Islam S, et al.
    Front Med (Lausanne), 2024;11:1407376.
    PMID: 39071085 DOI: 10.3389/fmed.2024.1407376
    INTRODUCTION: Global Cardiovascular disease (CVD) is still one of the leading causes of death and requires the enhancement of diagnostic methods for the effective detection of early signs and prediction of the disease outcomes. The current diagnostic tools are cumbersome and imprecise especially with complex diseases, thus emphasizing the incorporation of new machine learning applications in differential diagnosis.

    METHODS: This paper presents a new machine learning approach that uses MICE for mitigating missing data, the IQR for handling outliers and SMOTE to address first imbalance distance. Additionally, to select optimal features, we introduce the Hybrid 2-Tier Grasshopper Optimization with L2 regularization methodology which we call GOL2-2T. One of the promising methods to improve the predictive modelling is an Adaboost decision fusion (ABDF) ensemble learning algorithm with babysitting technique implemented for the hyperparameters tuning. The accuracy, recall, and AUC score will be considered as the measures for assessing the model.

    RESULTS: On the results, our heart disease prediction model yielded an accuracy of 83.0%, and a balanced F1 score of 84.0%. The integration of SMOTE, IQR outlier detection, MICE, and GOL2-2T feature selection enhances robustness while improving the predictive performance. ABDF removed the impurities in the model and elaborated its effectiveness, which proved to be high on predicting the heart disease.

    DISCUSSION: These findings demonstrate the effectiveness of additional machine learning methodologies in medical diagnostics, including early recognition improvements and trustworthy tools for clinicians. But yes, the model's use and extent of work depends on the dataset used for it really. Further work is needed to replicate the model across different datasets and samples: as for most models, it will be important to see if the results are generalizable to populations that are not representative of the patient population that was used for the current study.

  2. Jiao JY, Abdugheni R, Zhang DF, Ahmed I, Ali M, Chuvochina M, et al.
    Natl Sci Rev, 2024 Jul;11(7):nwae168.
    PMID: 39071100 DOI: 10.1093/nsr/nwae168
    Prokaryotes are ubiquitous in the biosphere, important for human health and drive diverse biological and environmental processes. Systematics of prokaryotes, whose origins can be traced to the discovery of microorganisms in the 17th century, has transitioned from a phenotype-based classification to a more comprehensive polyphasic taxonomy and eventually to the current genome-based taxonomic approach. This transition aligns with a foundational shift from studies focused on phenotypic traits that have limited comparative value to those using genome sequences. In this context, Bergey's Manual of Systematics of Archaea and Bacteria (BMSAB) and Bergey's International Society for Microbial Systematics (BISMiS) play a pivotal role in guiding prokaryotic systematics. This review focuses on the historical development of prokaryotic systematics with a focus on the roles of BMSAB and BISMiS. We also explore significant contributions and achievements by microbiologists, highlight the latest progress in the field and anticipate challenges and opportunities within prokaryotic systematics. Additionally, we outline five focal points of BISMiS that are aimed at addressing these challenges. In conclusion, our collaborative effort seeks to enhance ongoing advancements in prokaryotic systematics, ensuring its continued relevance and innovative characters in the contemporary landscape of genomics and bioinformatics.
  3. Zainal Ariffin SH, Megat Abdul Wahab R, Abdul Razak M, Yazid MD, Shahidan MA, Miskon A, et al.
    PeerJ, 2024;12:e17790.
    PMID: 39071131 DOI: 10.7717/peerj.17790
    BACKGROUND: Understanding human stem cell differentiation into osteoblasts and osteoclasts is crucial for bone regeneration and disease modeling. Numerous morphological techniques have been employed to assess this differentiation, but a comprehensive review of their application and effectiveness is lacking.

    METHODS: Guided by the PRISMA framework, we conducted a rigorous search through the PubMed, Web of Science and Scopus databases, analyzing 254 articles. Each article was scrutinized against pre-defined inclusion criteria, yielding a refined selection of 14 studies worthy of in-depth analysis.

    RESULTS: The trends in using morphological approaches were identified for analyzing osteoblast and osteoclast differentiation. The three most used techniques for osteoblasts were Alizarin Red S (mineralization; six articles), von Kossa (mineralization; three articles) and alkaline phosphatase (ALP; two articles) followed by one article on Giemsa staining (cell morphology) and finally immunochemistry (three articles involved Vinculin, F-actin and Col1 biomarkers). For osteoclasts, tartrate-resistant acid phosphatase (TRAP staining) has the highest number of articles (six articles), followed by two articles on DAPI staining (cell morphology), and immunochemistry (two articles with VNR, Cathepsin K and TROP2. The study involved four stem cell types: peripheral blood monocyte, mesenchymal, dental pulp, and periodontal ligament.

    CONCLUSION: This review offers a valuable resource for researchers, with Alizarin Red S and TRAP staining being the most utilized morphological procedures for osteoblasts and osteoclasts, respectively. This understanding provides a foundation for future research in this rapidly changing field.

    MeSH terms: Cell Differentiation*; Humans; Staining and Labeling/methods
  4. Bhowmik S, Mamun AA, Nordin N
    Front Public Health, 2024;12:1438907.
    PMID: 39071146 DOI: 10.3389/fpubh.2024.1438907
    MeSH terms: Agriculture; Humans; Occupational Health*; Safety Management*
  5. Wei LS, Adrian Susin AA, Tahiluddin AB, Kien LV, Wee W
    Heliyon, 2024 Jul 15;10(13):e33810.
    PMID: 39071570 DOI: 10.1016/j.heliyon.2024.e33810
    This study explores the beneficial effects of Auricularia auricula (AA) as a feed additive in promoting growth, digestive enzyme activities, antioxidative responses, heat tolerance, and disease resistance against Edwardsiella tarda in African catfish (Clarias gariepinus) farming. The application of feed additives is a hot topic in recent aquaculture studies aimed at promoting the growth and health of aquaculture species. After 8 weeks of feeding trial, the results of the present study revealed that fish-fed AA diets performed significantly better (p 
  6. Ashique S, Mohanto S, Ahmed MG, Mishra N, Garg A, Chellappan DK, et al.
    Heliyon, 2024 Jul 15;10(13):e34092.
    PMID: 39071627 DOI: 10.1016/j.heliyon.2024.e34092
    The microbiota-gut-brain axis (MGBA) represents a sophisticated communication network between the brain and the gut, involving immunological, endocrinological, and neural mediators. This bidirectional interaction is facilitated through the vagus nerve, sympathetic and parasympathetic fibers, and is regulated by the hypothalamic-pituitary-adrenal (HPA) axis. Evidence shows that alterations in gut microbiota composition, or dysbiosis, significantly impact neurological disorders (NDs) like anxiety, depression, autism, Parkinson's disease (PD), and Alzheimer's disease (AD). Dysbiosis can affect the central nervous system (CNS) via neuroinflammation and microglial activation, highlighting the importance of the microbiota-gut-brain axis (MGBA) in disease pathogenesis. The microbiota influences the immune system by modulating chemokines and cytokines, impacting neuronal health. Synbiotics have shown promise in treating NDs by enhancing cognitive function and reducing inflammation. The gut microbiota's role in producing neurotransmitters and neuroactive compounds, such as short-chain fatty acids (SCFAs), is critical for CNS homeostasis. Therapeutic interventions targeting the MGBA, including dietary modulation and synbiotic supplementation, offer potential benefits for managing neurodegenerative disorders. However, more in-depth clinical studies are necessary to fully understand and harness the therapeutic potential of the MGBA in neurological health and disease.
  7. Peng R, Razak RA, Halili SH
    Heliyon, 2024 Jul 15;10(13):e34234.
    PMID: 39071656 DOI: 10.1016/j.heliyon.2024.e34234
    Effectively incorporating technology into teaching and learning can be accomplished through the application of information and communication technology (ICT). Nonetheless, the extent to which ICT is employed in educational environments is heavily impacted by numerous factors. Teachers' ICT integration in China was investigated with the aim of identifying the specific factors that influence it. The research sample consists of 680 educators, and it highlights three primary elements that affect their adoption of ICT: attitudes, self-efficacy, and digital competence. Employing a partial least squares structural equation modeling (PLS-SEM) method, the results indicate that all three elements play a significant role in the effective integration of ICT by educators. The study also indicated the mediating effect of attitudes and digital competence. Furthermore, the research identified no substantial disparities in the factors based on sex or age, except for the correlation between self-efficacy and attitudes. By providing useful insights for the development of successful instructional designs that integrate ICT, this study contributed to the advancement and impact of educational technology.
  8. Zhining Z, Alli H, Ahmadipour M, Che Me R
    Heliyon, 2024 Jul 15;10(13):e34138.
    PMID: 39071662 DOI: 10.1016/j.heliyon.2024.e34138
    The integration of sustainable practices within manufacturing organizations has become a necessity. However, ensuring a competitive edge in the market remains pivotal for the success of these sustainability initiatives. This research introduces an approach to harmonize the influence of sustainability and agility within the product development process, enabling enterprises to pursue sustainable manufacturing while upholding robust market competitiveness. The significance of this study lies in its combined utilization of expert insights and mathematical techniques to gauge the components and sub-components of sustainability and agility, thereby enhancing the precision of assessment outcomes. This accomplishment was achieved through the application of a Weighted Fuzzy Assessment Method (WFAM) for evaluating both product sustainability and agility. Employing the Fuzzy Analytic Hierarchy Process (FAHP), the study assigned weights to elements and sub-elements. Subsequently, employing fuzzy logic based on these derived weights, the study assessed the sustainability and agility scores in the product development process. Demonstrating the effectiveness of this devised methodology, the research employed a multi-functional electric bicycle as a case study. The outcomes highlight the potential the proposed method in attaining the varied objectives of sustainability and agility in product development.
  9. Khairul Anuar AN, Muhammad Zamri NK, Mohd Yuzaidey MA, Su Nyun Pau S, Mustaffa NIH
    Data Brief, 2024 Aug;55:110683.
    PMID: 39071957 DOI: 10.1016/j.dib.2024.110683
    This article describes the abundance of phytoplankton community structures in Port Dickson, Negeri Sembilan and Pulau Tinggi, Johor during the Southwest and Northeast Monsoons and includes data from 48 selected sampling sites collected between July and December 2023. The seawater samples from 1-meter depth were obtained by using a Niskin water sampler, concentrated in a 50 ml centrifuge tube and immediately preserved with Lugol's iodine solution. The data include phytoplankton density (cell L-1), the total density of phytoplankton in each station, and the total number of genera obtained in every station. Additional data are presented, including chlorophyll-a concentration, as a proxy for biomass and photosynthetic active radiation. This article presents data on 30 genera, including unidentified genera, as well as the percentage of the main community group.
  10. Md Suhaimin MS, Ahmad Hijazi MH, Moung EG
    Data Brief, 2024 Aug;55:110663.
    PMID: 39071961 DOI: 10.1016/j.dib.2024.110663
    Sentiment analysis in the public security domain involves analysing public sentiment, emotions, opinions, and attitudes toward events, phenomena, and crises. However, the complexity of sarcasm, which tends to alter the intended meaning, combined with the use of bilingual code-mixed content, hampers sentiment analysis systems. Currently, limited datasets are available that focus on these issues. This paper introduces a comprehensive dataset constructed through a systematic data acquisition and annotation process. The acquisition process includes collecting data from social media platforms, starting with keyword searching, querying, and scraping, resulting in an acquired dataset. The subsequent annotation process involves refining and labelling, starting with data merging, selection, and annotation, ending in an annotated dataset. Three expert annotators from different fields were appointed for the labelling tasks, which produced determinations of sentiment and sarcasm in the content. Additionally, an annotator specialized in literature was appointed for language identification of each content. This dataset represents a valuable contribution to the field of natural language processing and machine learning, especially within the public security domain and for multilingual countries in Southeast Asia.
  11. Omar Z, P P Abdul Majeed A, Rosbi M, Ghazalli SA, Selamat H
    Data Brief, 2024 Aug;55:110667.
    PMID: 39071971 DOI: 10.1016/j.dib.2024.110667
    This dataset comprises oil palm fresh fruit bunch (FFB) images that may potentially be used in the study related to fruit ripeness detection via image processing. The FFB dataset was collected from palm oil plantations in Johor, Negeri Sembilan, and Perak, Malaysia. The data collection involved acquiring pictures of FFB from various angles and classifying them based on their ripeness level, categorised into five classes: damaged bunch, empty bunch, unripe, ripe, and overripe. An experienced grader carefully labelled each FFB image with the corresponding ground truth information. The dataset provides valuable insights into the colour variations of FFBs throughout their ripening process, which is essential for assessing oil quality. It includes observations on the external fruit colours as well as characteristics related to the presence of empty sockets in the FFB as a key indicator of ripeness. The reusability potential of this dataset is significant for researchers in the field of oil palm fruit classification and grading, which requires an extensive outdoor dataset that comprise FFB's both on the tree and on the ground. Our work enables the development and validation of machine learning pipelines for outdoor automated FFB grading. Furthermore, the dataset may also support studies to improve oil palm cultivation practices, enhance yield, and optimise oil quality.
  12. Hadi SH, Shaba TG, Madhi ZS, Darus M, Lupaş AA, Tchier F
    MethodsX, 2024 Dec;13:102842.
    PMID: 39071992 DOI: 10.1016/j.mex.2024.102842
    The study of holomorphic functions has been recently extended through the application of diverse techniques, among which quantum calculus stands out due to its wide-ranging applications across various scientific disciplines. In this context, we introduce a novel q-differential operator defined via the generalized binomial series, which leads to the derivation of new classes of quantum-convex (q-convex) functions. Several specific instances within these classes were explored in detail. Consequently, the boundary values of the Hankel determinants associated with these functions were analyzed. All graphical representations and computational analyses were performed using Mathematica 12.0.•These classes are defined by utilizing a new q-differential operator.•The coefficient values | a i | ( i = 2 , 3 , 4 ) are investigated.•Toeplitz determinants, such as the second T 2 ( 2 ) and the third T 3 ( 1 ) order inequalities, are calculated.
  13. Rehman SU, Sadek I, Huang B, Manickam S, Mahmoud LN
    MethodsX, 2024 Dec;13:102834.
    PMID: 39071997 DOI: 10.1016/j.mex.2024.102834
    The use of technology in healthcare is one of the most critical application areas today. With the development of medical applications, people's quality of life has improved. However, it is impractical and unnecessary for medium-risk people to receive specialized daily hospital monitoring. Due to their health status, they will be exposed to a high risk of severe health damage or even life-threatening conditions without monitoring. Therefore, remote, real-time, low-cost, wearable, and effective monitoring is ideal for this problem. Many researchers mentioned that their studies could use electrocardiogram (ECG) detection to discover emergencies. However, how to respond to discovered emergencies in household life is still a research gap in this field.•This paper proposes a real-time monitoring of ECG signals and sending them to the cloud for Sudden Cardiac Death (SCD) prediction.•Unlike previous studies, the proposed system has an additional emergency response mechanism to alert nearby community healthcare workers when SCD is predicted to occur.
  14. Abdulhameed EA, Rani KGA, AlGhalban FM, Abou Neel EA, Khalifa N, Khalil KA, et al.
    ACS Omega, 2024 Jul 23;9(29):31776-31788.
    PMID: 39072128 DOI: 10.1021/acsomega.4c02858
    Increased oxidative stress in bone cells is known to negatively alter favorable bone regeneration. This study aimed to develop a porous polycaprolactone (PCL) membrane incorporated with 25 wt % Vitamin C (PCL-Vit C) and compared it to the PCL membrane to control oxidative stress and enhance biomineralization in vitro. Both membranes were characterized using SEM-EDS, FTIR spectroscopy, and surface hydrophilicity. Vitamin C release was quantified colorimetrically. Assessments of the viability and attachment of human fetal osteoblast (hFOB 1.19) cells were carried out using XTT assay, SEM, and confocal microscopy, respectively. ROS generation and wound healing percentage were measured using flow cytometry and ImageJ software, respectively. Mineralization study using Alizarin Red in the presence or absence of osteogenic media was carried out to measure the calcium content. Alkaline phosphatase assay and gene expression of osteogenic markers (alkaline phosphatase (ALP), collagen Type I (Col1), runt-related transcription factor 2 (RUNX2), osteocalcin (OCN), and osteopontin (OPN)) were analyzed by real-time PCR. SEM images revealed smooth, fine, bead-free fibers in both membranes. The FTIR spectrum of pure vitamin C was replaced with peaks at 3436.05 and 2322.83 cm-1 in the PCL-Vit C membrane. Vitamin C release was detected at 15 min and 1 h. The PCL-Vit C membrane was hydrophilic, generated lower ROS, and showed significantly higher viability than the PCL membrane. Although both PCL and PCL-Vit C membranes showed similar cellular and cytoskeletal morphology, more cell clusters were evident in the PCL-Vit C membrane. Lower ROS level in the PCL-Vit C membrane displayed improved cell functionality as evidenced by enhanced cellular differentiation with more intense alizarin staining and higher calcium content, supported by upregulation of osteogenic markers ALP, Col1, and OPN even in the absence of osteogenic supplements. The presence of Vitamin C in the PCL-Vit C membrane may have mitigated oxidative stress in hFOB 1.19 cells, resulting in enhanced biomineralization facilitating bone regeneration.
  15. Eid N, Davamani F
    World J Gastrointest Oncol, 2024 Jul 15;16(7):2894-2901.
    PMID: 39072156 DOI: 10.4251/wjgo.v16.i7.2894
    Macroautophagy (hereafter referred to as autophagy) is a prosurvival mechanism for the clearance of damaged cellular components, specifically related to exposure to various stressors such as starvation, excessive ethanol intake, and chemotherapy. This editorial reviews and comments on an article by Zhao et al, to be published in World J Gastrointestinal Oncology in 2024. Based on various molecular biology methodologies, they found that human β-defensin-1 reduced the proliferation of colon cancer cells, which was associated with the inhibition of the mammalian target of rapamycin, resulting in autophagy activation. The activation of autophagy is evidenced by increased levels of Beclin1 and LC3II/I proteins and mediated by the upregulation of long non-coding RNA TCONS_00014506. Our study discusses the impact of autophagy activation and mechanisms of autophagy, including autophagic flux, on cancer cells. Additionally, we emphasize the importance of describing the detailed methods for isolating long noncoding RNAs TCONS_00014506. Our review will benefit the scientific community and improve the overall clarity of the paper.
  16. Liu P, Yuan H, Lu Y, Gao Z
    Front Physiol, 2024;15:1424216.
    PMID: 39072216 DOI: 10.3389/fphys.2024.1424216
    INTRODUCTION: This study aimed to evaluate the effects of varied resistance training modalities on physical fitness components, body composition, maximal strength assessed by one-repetition maximum (1RM), isokinetic muscle functions of the shoulder and knee joints, and biomechanical properties of core muscles.

    METHODS: Forty participants were randomly assigned to four groups: control group (CG, n = 10), compound set training group (CSG, n = 10), pyramid set training group (PSG, n = 10), and superset training group (SSG, n = 10). Excluding the CG, the other three groups underwent an 8-week resistance training program, three sessions per week, at 60%-80% of 1RM intensity for 60-90 min per session. Assessments included body composition, physical fitness components, 1RM, isokinetic muscle functions, and biomechanical properties (muscle frequency, stiffness, etc.) of the rectus abdominis and external oblique muscles.

    RESULTS: The PSG demonstrated the most significant improvement in relative peak torque during isokinetic testing of the shoulder and knee joints. Compared to the CG, all exercise groups exhibited positive effects on back strength, sprint performance, 1RM, and core muscle biomechanics. Notably, the PSG showed superior enhancement in external oblique stiffness. However, no significant differences were observed among the exercise groups for rectus abdominis biomechanical properties.

    DISCUSSION: Structured resistance training effectively improved maximal strength, functional performance, and core muscle biomechanics. The pyramidal training modality conferred specific benefits for isokinetic muscle functions and external oblique stiffness, suggesting its efficacy in enhancing force production capabilities and core stability.

  17. Young PJ, Al-Fares A, Aryal D, Arabi YM, Ashraf MS, Bagshaw SM, et al.
    Crit Care Resusc, 2024 Jun;26(2):87-94.
    PMID: 39072241 DOI: 10.1016/j.ccrj.2024.03.004
    BACKGROUND: The effect of conservative vs. liberal oxygen therapy on 90-day in-hospital mortality in adults with hypoxic ischaemic encephalopathy (HIE) following a cardiac arrest who are receiving invasive mechanical ventilation in the intensive care unit (ICU) is uncertain.

    OBJECTIVE: To summarise the protocol and statistical analysis plan for the Mega-ROX HIE trial.

    DESIGN SETTING AND PARTICIPANTS: Mega-ROX HIE is an international randomised clinical trial that will be conducted within an overarching 40,000-participant registry-embedded clinical trial comparing conservative and liberal ICU oxygen therapy regimens. We expect to enrol approximately 4000 participants with suspected HIE following a cardiac arrest who are receiving invasive mechanical ventilation in the ICU.

    MAIN OUTCOME MEASURES: The primary outcome is in-hospital all-cause mortality up to 90 days from the date of randomisation. Secondary outcomes include duration of survival, duration of mechanical ventilation, ICU length of stay, hospital length of stay, and the proportion of participants discharged home.

    RESULTS AND CONCLUSIONS: Mega-ROX HIE will compare the effect of conservative vs. liberal oxygen therapy regimens on day-90 in-hospital mortality in adults in the ICU with suspected HIE following a cardiac arrest. The protocol and planned analyses are reported here to mitigate analysis bias.

    TRIAL REGISTRATION: Australian and New Zealand Clinical Trials Registry (ACTRN 12620000391976).

  18. Ding L, Wang G, Wang J, Peng Y, Cai S, Khan SU, et al.
    J Control Release, 2024 Aug;372:43-58.
    PMID: 38866243 DOI: 10.1016/j.jconrel.2024.06.022
    Chronic infections often involve biofilm-based bacteria, in which the biofilm results in significant resistance against antimicrobial agents and prevents eradication of the infection. The physicochemical barrier presented by the biofilm matrix is a major impediment to the delivery of many antibiotics. Previously, PEGylation has been shown to improve antibiotic penetration into biofilms in vitro. In these studies, PEGylating tobramycin was investigated both in vitro and in vivo. Two distinct PEGylated tobramycin molecules were synthesized (mPEG-SA-Tob and mPEG-AA-Tob). Then, in a P. aeruginosa biofilm in vitro model, we found that mPEG-SA-Tob can operate as a prodrug and showed 7 times more effectiveness than tobramycin (MIC80: 14 μM vs.100 μM). This improved biofilm eradication is attributable to the fact that mPEG-SA-Tob can aid tobramycin to penetrate through the biofilm and overcome the alginate-mediated antibiotic resistance. Finally, we used an in vivo biofilm-based chronic pulmonary infection rat model to confirm the therapeutic impact of mPEG-SA-Tob on biofilm-based chronic lung infection. mPEG-SA-Tob has a better therapeutic impact than tobramycin in that it cannot only stop P. aeruginosa from multiplying in the lungs but can also reduce inflammation caused by infections and prevent a recurrence infection. Overall, our findings show that PEGylated tobramycin is an effective treatment for biofilm-based chronic lung infections.
    MeSH terms: Animals; Male; Microbial Sensitivity Tests; Rats, Sprague-Dawley; Rats
  19. Ong JEX, Blum IR
    Prim Dent J, 2024 Jun;13(2):58-64.
    PMID: 38888073 DOI: 10.1177/20501684241249558
    This clinical case report demonstrates the use of the Dahl Concept in the management of the repeated dislodgement of a posterior full coverage crown associated with a reduced restorative space. The described technique harnesses the addition of resin composite and a temporarily cemented provisional full coverage crown to create sufficient restorative space for the cementation of a definitive posterior full coverage crown restoration at the six-month review.
    MeSH terms: Cementation; Crowns*; Female; Humans
  20. Ferdowsi M, Hasan MM, Habib W
    Comput Methods Programs Biomed, 2024 Sep;254:108289.
    PMID: 38905988 DOI: 10.1016/j.cmpb.2024.108289
    BACKGROUND AND OBJECTIVE: Cardiovascular disease (CD) is a major global health concern, affecting millions with symptoms like fatigue and chest discomfort. Timely identification is crucial due to its significant contribution to global mortality. In healthcare, artificial intelligence (AI) holds promise for advancing disease risk assessment and treatment outcome prediction. However, machine learning (ML) evolution raises concerns about data privacy and biases, especially in sensitive healthcare applications. The objective is to develop and implement a responsible AI model for CD prediction that prioritize patient privacy, security, ensuring transparency, explainability, fairness, and ethical adherence in healthcare applications.

    METHODS: To predict CD while prioritizing patient privacy, our study employed data anonymization involved adding Laplace noise to sensitive features like age and gender. The anonymized dataset underwent analysis using a differential privacy (DP) framework to preserve data privacy. DP ensured confidentiality while extracting insights. Compared with Logistic Regression (LR), Gaussian Naïve Bayes (GNB), and Random Forest (RF), the methodology integrated feature selection, statistical analysis, and SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) for interpretability. This approach facilitates transparent and interpretable AI decision-making, aligning with responsible AI development principles. Overall, it combines privacy preservation, interpretability, and ethical considerations for accurate CD predictions.

    RESULTS: Our investigations from the DP framework with LR were promising, with an area under curve (AUC) of 0.848 ± 0.03, an accuracy of 0.797 ± 0.02, precision at 0.789 ± 0.02, recall at 0.797 ± 0.02, and an F1 score of 0.787 ± 0.02, with a comparable performance with the non-privacy framework. The SHAP and LIME based results support clinical findings, show a commitment to transparent and interpretable AI decision-making, and aligns with the principles of responsible AI development.

    CONCLUSIONS: Our study endorses a novel approach in predicting CD, amalgamating data anonymization, privacy-preserving methods, interpretability tools SHAP, LIME, and ethical considerations. This responsible AI framework ensures accurate predictions, privacy preservation, and user trust, underscoring the significance of comprehensive and transparent ML models in healthcare. Therefore, this research empowers the ability to forecast CD, providing a vital lifeline to millions of CD patients globally and potentially preventing numerous fatalities.

    MeSH terms: Data Anonymization; Machine Learning*; Algorithms; Artificial Intelligence*; Bayes Theorem; Confidentiality; Female; Humans; Male; Middle Aged; Logistic Models; Risk Assessment/methods; Privacy
External Links