Disease Prediction Research Articles

Genetic Architectures of Adolescent Depression Trajectories in 2 Longitudinal Population Cohorts.

Adolescent depression is characterized by diverse symptom trajectories over time and has a strong genetic influence. Research has determined genetic overlap between depression and other psychiatric conditions; investigating the shared genetic architecture of heterogeneous depression trajectories is crucial for understanding disease etiology, prediction, and early intervention. To investigate univariate and multivariate genetic risk for adolescent depression trajectories and assess generalizability across ancestries. This cohort study entailed longitudinal growth modeling followed by polygenic risk score (PRS) association testing for individual and multitrait genetic models. Two longitudinal cohorts from the US and UK were used: the Adolescent Brain and Cognitive Development (ABCD; N = 11 876) study and the Avon Longitudinal Study of Parents and Children (ALSPAC; N = 8787) study. Included were adolescents with genetic information and depression measures at up to 8 and 4 occasions, respectively. Study data were analyzed January to July 2023. Trajectories were derived from growth mixture modeling of longitudinal depression symptoms. PRSs were computed for depression, anxiety, neuroticism, bipolar disorder, schizophrenia, attention-deficit/hyperactivity disorder, and autism in European ancestry. Genomic structural equation modeling was used to build multitrait genetic models of psychopathology followed by multitrait PRS. Depression PRSs were computed in African, East Asian, and Hispanic ancestries in the ABCD cohort only. Association testing was performed between all PRSs and trajectories for both cohorts. A total sample size of 14 112 adolescents (at baseline: mean [SD] age, 10.5 [0.5] years; 7269 male sex [52%]) from both cohorts were included in this analysis. Distinct depression trajectories (stable low, adolescent persistent, increasing, and decreasing) were replicated in the ALSPAC cohort (6096 participants; 3091 female [51%]) and ABCD cohort (8016 participants; 4274 male [53%]) between ages 10 and 17 years. Most univariate PRSs showed significant uniform associations with persistent trajectories, but fewer were significantly associated with intermediate (increasing and decreasing) trajectories. Multitrait PRSs-derived from a hierarchical factor model-showed the strongest associations for persistent trajectories (ABCD cohort: OR, 1.46; 95% CI, 1.26-1.68; ALSPAC cohort: OR, 1.34; 95% CI, 1.20-1.49), surpassing the effect size of univariate PRS in both cohorts. Multitrait PRSs were associated with intermediate trajectories but to a lesser extent (ABCD cohort: hierarchical increasing, OR, 1.27; 95% CI, 1.13-1.43; decreasing, OR, 1.23; 95% CI, 1.09-1.40; ALSPAC cohort: hierarchical increasing, OR, 1.16; 95% CI, 1.04-1.28; decreasing, OR, 1.32; 95% CI, 1.18-1.47). Transancestral genetic risk for depression showed no evidence for association with trajectories. Results of this cohort study revealed a high multitrait genetic loading of persistent symptom trajectories, consistent across traits and cohorts. Variability in univariate genetic association with intermediate trajectories may stem from environmental factors. Multitrait genetics may strengthen depression prediction models, but more diverse data are needed for generalizability.

Choriocapillaris reduction accurately discriminates against early-onset Alzheimer's disease.

This study addresses the urgent need for non-invasive early-onset Alzheimer's disease (EOAD) prediction. Using optical coherence tomography angiography (OCTA), we present a choriocapillaris model sensitive to EOAD, correlating with serum biomarkers. Eighty-four EOAD patients and 73 controls were assigned to swept-source OCTA (SS-OCTA) or the spectral domain OCTA (SD-OCTA) cohorts. Our hypothesis on choriocapillaris predictive potential in EOAD was tested and validated in these two cohorts. Both cohorts revealed diminished choriocapillaris signals, demonstrating the highest discriminatory capability (area under the receiver operating characteristic curve: SS-OCTA 0.913, SD-OCTA 0.991; P<0.001). A sparser SS-OCTA choriocapillaris correlated with increased serum amyloid beta (Aβ)42, Aβ42/40, and phosphorylated tau (p-tau)181 levels (all P<0.05). Apolipoprotein E status did not affect choriocapillaris measurement. The choriocapillaris, observed in both cohorts, proves sensitive to EOAD diagnosis, and correlates with serum Aβ and p-tau181 levels, suggesting its potential as a diagnostic tool for identifying and tracking microvascular changes in EOAD. Optical coherence tomography angiography may be applied for non-invasive screening of Alzheimer's disease (AD). Choriocapillaris demonstrates high sensitivity and specificity for early-onset AD diagnosis. Microvascular dynamics abnormalities are associated with AD.

Comparative Analysis of Machine Learning Algorithms for Liver Disease Prediction: SVM, Logistic Regression, and Decision Tree

Comparative Analysis of Machine Learning Algorithms for Liver Disease Prediction: SVM, Logistic Regression, and Decision Tree

HSLE: A Hybrid Ensemble Classifier for Prediction of Heart Disease

Background: Detecting heart disease in a timely manner is vital for preventing its progression, as it is the primary cause of death across the globe. Machine learning has the potential to enhance diagnostic accuracy and enable better clinical decision-making. A machine learningpowered hybrid system for diagnosing heart disease may provide a better optimal solution for heart disease prediction. Objective: The overarching objectives include accuracy improvement, enhanced classification reliability, and the development of high-performance prediction models for heart disease. These objectives indicate a commitment to advancing methodologies and models in the field of machine learning and data science, particularly within the domain of healthcare and disease prediction. Method: The proposed system was developed using the Cleveland dataset that was preprocessed and analyzed using Recursive Feature Elimination with Cross-Validation (RFECV) and Least Absolute Shrinkage and Selection Operator (LASSO) feature extraction techniques. Further, a hybrid feature selection approach using RFECV and K-Best has been proposed for feature selection. Eight machine learning classifiers such as Multilayer Perceptron (MLP), Random Forest (RF), Extreme Gradient Boosting (XGB), K-Nearest Neighbours (KNN), Extra Tree (ET), Support Vector Machine (SVC), Adaboost, Decision Tree (DT) were utilized, and the performance of the system was measured in terms of various metrics. Result: The results showed that the proposed HSLE algorithm with hybrid feature selection led to the highest overall accuracy of 98.76%. Conclusion: As mentioned, the main cause of adult death worldwide is chronic disease. Early detection can stop the condition from getting worse. Our research presents an innovative hybrid machine-learning approach designed to forecast heart disease.

Lipidome is a valuable tool for the severity prediction of coronavirus disease 2019

ObjectiveTo describe the lipid metabolic profile of different patients with coronavirus disease 2019 (COVID-19) and contribute new evidence on the progression and severity prediction of COVID-19.MethodsThis case–control study was conducted in Peking University Third Hospital, China. The laboratory-confirmed COVID-19 patients aged ≥18 years old and diagnosed as pneumonia from December 2022 to January 2023 were included. Serum lipids were detected. The discrimination ability was calculated with the area under the curve (AUC). A random forest (RF) model was conducted to determine the significance of different lipids.ResultsTotally, 44 COVID-19 patients were enrolled with 16 mild and 28 severe patients. The top 5 super classes were triacylglycerols (TAG, 55.9%), phosphatidylethanolamines (PE, 10.9%), phosphatidylcholines (PC, 6.8%), diacylglycerols (DAG, 5.9%) and free fatty acids (FFA, 3.6%) among the 778 detected lipids from the serum of COVID-19 patients. Certain lipids, especially lysophosphatidylcholines (LPCs), turned to have significant correlations with certain immune/cytokine indexes. Reduced level of LPC 20:0 was observed in severe patients particularly in acute stage. The AUC of LPC 20:0 reached 0.940 in discriminating mild and severe patients and 0.807 in discriminating acute and recovery stages in the severe patients. The results of RF models also suggested the significance of LPCs in predicting the severity and progression of COVID-19.ConclusionLipids probably have the potential to differentiate and forecast the severity, progression, and clinical outcomes of COVID-19 patients, with implications for immune/inflammatory responses. LPC 20:0 might be a potential target in predicting the progression and outcome and the treatment of COVID-19.

Neonatal somatic oxygenation and perfusion assessment using near-infrared spectroscopy : Part of the series on near-infrared spectroscopy by the European Society of Paediatric Research Special Interest Group "Near-Infrared Spectroscopy".

In this narrative review, we summarize the current knowledge and applications of somatic near-infrared spectroscopy (NIRS), with a focus on intestinal, renal, limb, and multi-site applications in neonates. Assessing somatic oxygenation at various body locations in neonates may aid in the understanding of underlying pathophysiology of organ injury. Considering cerebral autoregulation may be active to protect the brain during systemic circulatory failure, peripheral somatic oxygenation may potentially provide an early indication of neonatal cardiovascular failure and ultimate hypoxemic injury to vital organs including the brain. Certain intestinal oxygenation patterns appear to be associated with the onset and course of necrotizing enterocolitis, whereas impaired renal oxygenation may indicate the onset of acute kidney injury after various types of hypoxic events. Peripheral muscle oxygenation measured at a limb may be particularly effective in the early prediction of shock in neonates. Using multi-site NIRS may complement current approaches and clinical investigations to alert for neonatal tissue hypoxemia, and potentially even guide management. However, somatic NIRS has its inherent limitations in regard to accuracy. Interpretation of organ-specific values can also be challenging. Last, currently there are limited prospective intervention studies, and clinical benefits need to be examined further, after the clarification of critical threshold-values. IMPACT: The assessment of somatic oxygenation using NIRS may contribute to the prediction of specific diseases in hemodynamically challenged neonates. Furthermore, it may give early warning signs for impending cardiovascular failure, and impaired cerebral circulation and oxygenation. We present a comprehensive overview of the literature on applications of NIRS to various somatic areas, with a focus on its potential clinical applicability, including future research directions. This paper will enable prospective standardized studies, and multicenter collaboration to obtain statistical power, likely to advance the field.

Predictive genome-wide association and economic analyses of pooled DNA in a commercial feedlot setting

Understanding the genetic profile of cattle allows producers to make predictions of performance and disease. However, it is often fiscally and logistically unfeasible for commercial cattle producers to individually genotype animals. As such, our objec­tive was to utilize prospectively-pooled DNA from commercial feedlot cattle for a genome-wide association study (GWAS) of ec­onomically-important carcass traits and measure the economic efficiency of this technique.

Efficient Machine Learning Techniques to Classifying Cardiovascular Disease and Improve Prediction Analysis

Cardiovascular Disease (CVD) account for a large portion of the global health burden and are one of the main causes of decease worldwide. In the classification and forecasting of CVDs, Machine Learning (ML) techniques have demonstrated encouraging outcomes. In this research report, a comparative analysis of classification and prediction models for CVD is presented, including both linear and ensemble ML approaches. The paper compares ensemble models like Catboost, Histogram Gradient Boosting Machine (HGBM), and Extra Trees against linear models like Gaussian Nave Bayes, SVM, and KNN. The objective is to identify the most effective CVD prediction model by assessing its performance through accuracy, precision, sensitivity and F1 score as key evaluation metrics. Moreover, results show that ensemble models outperform linear models using advanced techniques such as boosting and histogram-based algorithms. The results underscore the critical role ensemble models play in accurately diagnosing and predicting cardiovascular disease and provide important new information to researchers and healthcare providers. Using these models has the potential to significantly improve patient outcomes and health management by enabling early detection and intervention

International Classification of Diseases Prediction from MIMIIC-III Clinical Text Using Pre-Trained ClinicalBERT and NLP Deep Learning Models Achieving State of the Art

International Classification of Diseases Prediction from MIMIIC-III Clinical Text Using Pre-Trained ClinicalBERT and NLP Deep Learning Models Achieving State of the Art

Precision healthcare: A deep dive into machine learning algorithms and feature selection strategies for accurate heart disease prediction

This paper presents a comprehensive exploration of machine learning algorithms (MLAs) and feature selection techniques for accurate heart disease prediction (HDP) in modern healthcare. By focusing on diverse datasets encompassing various challenges, the research sheds light on optimal strategies for early detection. MLAs such as Decision Trees (DT), Random Forests (RF), Support Vector Machines (SVM), Gaussian Naive Bayes (NB), and others were studied, with precision and recall metrics emphasized for robust predictions. Our study addresses challenges in real-world data through data cleaning and one-hot encoding, enhancing the integrity of our predictive models. Feature extraction techniques—Recursive Feature Extraction (RFE), Principal Component Analysis (PCA), and univariate feature selection—play a crucial role in identifying relevant features and reducing data dimensionality. Our findings showcase the impact of these techniques on improving prediction accuracy. Optimized models for each dataset have been achieved through grid search hyperparameter tuning, with configurations meticulously outlined. Notably, a remarkable 99.12 % accuracy was achieved on the first Kaggle dataset, showcasing the potential for accurate HDP. Model robustness across diverse datasets was highlighted, with caution against overfitting. The study emphasizes the need for validation of unseen data and encourages ongoing research for generalizability. Serving as a practical guide, this research aids researchers and practitioners in HDP model development, influencing clinical decisions and healthcare resource allocation. By providing insights into effective algorithms and techniques, the paper contributes to reducing heart disease-related morbidity and mortality, supporting the healthcare community's ongoing efforts.

Deep learning imaging phenotype can classify metabolic syndrome and is predictive of cardiometabolic disorders

BackgroundCardiometabolic disorders pose significant health risks globally. Metabolic syndrome, characterized by a cluster of potentially reversible metabolic abnormalities, is a known risk factor for these disorders. Early detection and intervention for individuals with metabolic abnormalities can help mitigate the risk of developing more serious cardiometabolic conditions. This study aimed to develop an image-derived phenotype (IDP) for metabolic abnormality from unenhanced abdominal computed tomography (CT) scans using deep learning. We used this IDP to classify individuals with metabolic syndrome and predict future occurrence of cardiometabolic disorders.MethodsA multi-stage deep learning approach was used to extract the IDP from the liver region of unenhanced abdominal CT scans. In a cohort of over 2,000 individuals the IDP was used to classify individuals with metabolic syndrome. In a subset of over 1,300 individuals, the IDP was used to predict future occurrence of hypertension, type II diabetes, and fatty liver disease.ResultsFor metabolic syndrome (MetS) classification, we compared the performance of the proposed IDP to liver attenuation and visceral adipose tissue area (VAT). The proposed IDP showed the strongest performance (AUC 0.82) compared to attenuation (AUC 0.70) and VAT (AUC 0.80). For disease prediction, we compared the performance of the IDP to baseline MetS diagnosis. The models including the IDP outperformed MetS for type II diabetes (AUCs 0.91 and 0.90) and fatty liver disease (AUCs 0.67 and 0.62) prediction and performed comparably for hypertension prediction (AUCs of 0.77).ConclusionsThis study demonstrated the superior performance of a deep learning IDP compared to traditional radiomic features to classify individuals with metabolic syndrome. Additionally, the IDP outperformed the clinical definition of metabolic syndrome in predicting future morbidities. Our findings underscore the utility of data-driven imaging phenotypes as valuable tools in the assessment and management of metabolic syndrome and cardiometabolic disorders.

A Method of Skin Disease Detection Using Image Processing and Machine Learning

Skin conditions are prevalent health issues globally. The dangers posed by these infections are unseen, leading to physical discomfort and mental anguish. In severe cases, skin diseases can even progress to skin cancer. Consequently, the identification of skin diseases from clinical images remains a significant challenge in medical image analysis. Furthermore, manual diagnosis by medical professionals is time-consuming and subjective. Therefore, there is a need for automated skin disease prediction to expedite treatment planning for both patients and dermatologists. These studies presents a digital hair removal method and de-blurs or denoise the images. To extract essential patterns from the skin imagesand statistical features techniques are utilized. Two efficient machine learning algorithms, namely Support Vector Machine (SVM), and Convolutional Neural Network (CNN) classifiers, are employed with the extracted features to accurately classify skin images as melanoma (MEL), melanocytic nevus (NV), basal cell carcinoma (BCC), actinic keratosis (AK), benign keratosis (BKL), dermatofibroma (DF), vascular lesion (VASC), and Squamous cell carcinoma (SCC). The models are validated using the ISIC 2019 challenge and HAM10000 datasets, with SVM demonstrating slightly superior performance compared to the other classifiers. Furthermore, a comparison with state-of-the-art methods is conducted to evaluate the effectiveness of the proposed approach.

Prediction and causal inference of cardiovascular and cerebrovascular diseases based on lifestyle questionnaires

Cardiovascular and cerebrovascular diseases (CCVD) are prominent mortality causes in Japan, necessitating effective preventative measures, early diagnosis, and treatment to mitigate their impact. A diagnostic model was developed to identify patients with ischemic heart disease (IHD), stroke, or both, using specific health examination data. Lifestyle habits affecting CCVD development were analyzed using five causal inference methods. This study included 473,734 patients aged ≥ 40 years who underwent specific health examinations in Kanazawa, Japan between 2009 and 2018 to collect data on basic physical information, lifestyle habits, and laboratory parameters such as diabetes, lipid metabolism, renal function, and liver function. Four machine learning algorithms were used: Random Forest, Logistic regression, Light Gradient Boosting Machine, and eXtreme-Gradient-Boosting (XGBoost). The XGBoost model exhibited superior area under the curve (AUC), with mean values of 0.770 (± 0.003), 0.758 (± 0.003), and 0.845 (± 0.005) for stroke, IHD, and CCVD, respectively. The results of the five causal inference analyses were summarized, and lifestyle behavior changes were observed after the onset of CCVD. A causal relationship from ‘reduced mastication’ to ‘weight gain’ was found for all causal species theory methods. This prediction algorithm can screen for asymptomatic myocardial ischemia and stroke. By selecting high-risk patients suspected of having CCVD, resources can be used more efficiently for secondary testing.

Prevalence of human respiratory pathogens and associated mucosal cytokine levels in young children and adults: a cross-sectional observational study in the Netherlands during the winter of 2012/2013.

Respiratory pathogens can cause severe disease and even death, especially in the very young and very old. Studies investigating their prevalence often focus on individuals presenting to healthcare providers with symptoms. However, the design of prevention strategies, e.g. which target groups to vaccinate, will benefit from knowledge on the prevalence of, risk factors for and host response to these pathogens in the general population. In this study, upper respiratory samples (n=1311) were collected cross-sectionally during winter from 11- and 24-month old children, their parents, and adults ≥60 years of age that were recruited irrespective of seeking medical care. Almost all children, approximately two thirds of parents and a quarter of older adults tested positive for at least one pathogen, often in the absence of symptoms. Viral interference was evident for the combination of rhinovirus and respiratory syncytial virus. Attending childcare facilities and having siblings associated with increased pathogen counts in children. On average, children showed increased levels of mucosal cytokines compared to parents and especially pro-inflammatory molecules associated with the presence of symptoms. These findings may guide further research into transmission patterns of respiratory pathogens and assist in determining the most appropriate strategies for the prediction and prevention of disease.

Multidimensional biomarker approach integrating tumor markers, inflammatory indicators, and disease activity indicators may improve prediction of rheumatoid arthritis-associated interstitial lung disease.

Rheumatoid arthritis (RA) often leads to interstitial lung disease (ILD), significantly affecting patient outcomes. This study explored the diagnostic accuracy of a multi-biomarker approach to offer a more efficient and accessible diagnostic strategy for RA-associated ILD (RA-ILD). Patients diagnosed with RA, with or without ILD, at Beijing Tiantan Hospital from October 2019 to October 2023 were analyzed. A total of 125 RA patients were included, with 76 diagnosed with RA-ILD. The study focused on three categories of indicators: tumor markers, inflammatory indicators, and disease activity measures. The heatmap correlation analysis was employed to analyze the correlation among these indicators. Logistic regression was used to determine odds ratios (OR) for indicators linked to RA-ILD risk. Receiver-operating characteristic (ROC) curve analysis was employed to evaluate the diagnostic potential of these indicators for RA-ILD. The results of logistic regression analysis showed that tumor markers (carbohydrate antigen 19-9 (CA19-9), carbohydrate antigen 125 (CA125), and cytokeratin 19 fragment (CYFRA21-1)), as well as inflammatory indicators (neutrophil, neutrophil-to-lymphocyte ratio (NLR), platelet, C-reactive protein (CRP)) and disease activity measures (disease activity score-28-CRP (DAS28-CRP), rheumatoid factor (RF), and anti-cyclic peptide containing citrulline (anti-CCP)), were significantly associated with RA-ILD. The correlation coefficients among these indicators were relatively low. Notably, the combination indicator 4, which integrated the aforementioned three categories of biomarkers, demonstrated improved diagnostic accuracy with an AUC of 0.857. The study demonstrated that combining tumor markers, inflammatory indicators, and disease activity measures significantly enhanced the prediction of RA-ILD. Key Points • Multidimensional strategy: Integrated tumor markers, inflammatory indicators, and disease activity measures to enhance early detection of rheumatoid arthritis-associated interstitial lung disease (RA-ILD). • Diagnostic accuracy: Employed heatmap correlation and logistic regression, identifying significant associations and improving diagnostic accuracy with a multidimensional biomarker combination. • Superior performance: The combined multidimensional biomarker strategy demonstrated higher diagnostic precision compared to individual or dual-category indicators. • Clinical relevance: Offers a promising, accessible approach for early detection of RA-ILD in clinical settings, potentially improving patient outcomes.

Systematic literature review to inform the EULAR recommendations for the use of imaging in crystal-induced arthropathies in clinical practice

ObjectiveTo summarise current data regarding the use of imaging in crystal-induced arthropathies (CiAs) informing a European Alliance of Associations for Rheumatology task force.MethodsWe performed four systematic searches in Embase, Medline...

MediBot For Symptoms-Driven Disease Prediction and Treatment Suggestions

To promote awareness regarding healthcare for underprivileged and inactive individuals globally, post-pandemic scenarios such as COVID-19 have led to hesitancy in seeking medical attention for minor ailments like fever and cold. This study presents a groundbreaking solution: the Medibot, which offers disease prediction based on symptoms and provides treatment recommendations through text and audio input and output. The Medibot assists users anytime, anywhere with its 24/7 services, saving time. Utilizing Deep Learning, it is capable of making decisions akin to humans and is also cost-effective for users. Keywords— “Health care”, “MediBot”, “Diseases prediction”, “Treatment recommendations”, “Deep learning”, “Cost-efficient”

Identifying autism spectrum disorder from multi-modal data with privacy-preserving

The application of deep learning models to precision medical diagnosis often requires the aggregation of large amounts of medical data to effectively train high-quality models. However, data privacy protection mechanisms make it difficult to perform medical data collection from different medical institutions. In autism spectrum disorder (ASD) diagnosis, automatic diagnosis using multimodal information from heterogeneous data has not yet achieved satisfactory performance. To address the privacy preservation issue as well as to improve ASD diagnosis, we propose a deep learning framework using multimodal feature fusion and hypergraph neural networks for disease prediction in federated learning (FedHNN). By introducing the federated learning strategy, each local model is trained and computed independently in a distributed manner without data sharing, allowing rapid scaling of medical datasets to achieve robust and scalable deep learning predictive models. To further improve the performance with privacy preservation, we improve the hypergraph model for multimodal fusion to make it suitable for autism spectrum disorder (ASD) diagnosis tasks by capturing the complementarity and correlation between modalities through a hypergraph fusion strategy. The results demonstrate that our proposed federated learning-based prediction model is superior to all local models and outperforms other deep learning models. Overall, our proposed FedHNN has good results in the work of using multi-site data to improve the performance of ASD identification.

Smart livestock management: integrating IoT for cattle health diagnosis and disease prediction through machine learning

Cattle diseases can significantly impact on livestock health and agricultural productivity is substantial. Timely detection and prognosis of these diseases are essential for prompt interventions and preventing their spread within the herd. This study delved into employing machine learning models to anticipate cattle diseases based on relevant parameters. These parameters encompass milk fever, milk clots, milk watery, milk flake, blisters, lameness, stomach pain, gaseous stomach, dehydration, diarrhea, vomiting, abdominal issues, and alkalosis. A dataset of 2,000 samples from diverse cattle populations was amassed, each tagged with the presence or absence of specific diseases. The primary goal was to compare the efficacy of five well-known machine learning models: Naïve Bayes multinomial (NBM), lazy-IBk, partial tree (PART), random forest (RF), and support vector machine (SVM). The findings underscored the consistent superiority of RF in comparison to the other models, boasting the highest accuracy in predicting cattle diseases. The RF model exhibited an accuracy rate of 88% on the test dataset. This achievement can be ascribed to its capacity to handle intricate interactions among input features and mitigate over fitting through ensemble learning. These insights can furnish valuable information about early indicators and risk factors associated with diverse cattle diseases.

A parameterised model for link prediction using node centrality and similarity measure based on graph embedding

Link prediction is a crucial aspect of graph machine learning, with applications as diverse as disease prediction, social network recommendations, and drug discovery. It involves the prediction of potential new links between nodes within a network. Despite its importance, current models for link prediction exhibit notable limitations. Graph Convolutional Networks have shown high efficiency in link prediction across various datasets. However, they face significant challenges when applied to short-path networks and ego networks, resulting in poor performance. This issue represents a critical area of concern that our work seeks to address. This paper introduces the Node Centrality and Similarity Based Parameterised Model (NCSM), a novel method for link prediction tasks. NCSM uniquely integrates node centrality and similarity measures as edge features in a customised Graph Neural Network (GNN) layer, effectively leveraging the topological information of large networks. This model represents the first parameterised GNN-based link prediction model that considers topological information. The proposed model was evaluated on five benchmark graph datasets, each comprising thousands of nodes and edges. Experimental results highlight NCSM's superiority over existing state-of-the-art models like Graph Convolutional Networks and Variational Graph Autoencoder, as it outperforms them across various metrics and datasets. This exceptional performance can be attributed to NCSM's innovative integration of node centrality, similarity measures, and its efficient use of topological information.