Jul 2026· IDEALIS : InDonEsiA journaL Information System· Vol 9, pp. 214-223· 0 citations· 21 references
TL;DR
Findings indicate that deep learning approaches, particularly CNN-LSTM, are more effective in capturing the complex dynamics of urban air pollution data and have strong potential to support air quality forecasting and pollution control systems in Jakarta.
Abstract
Increasing air pollution in urban areas, particularly in DKI Jakarta, has made a reliable air quality prediction system increasingly essential for environmental control and public health risk management. The Air Quality Index (AQI) exhibits complex and fluctuating patterns, requiring forecasting methods capable of capturing both linear and non-linear. This study aims to conduct a comparative analysis of statistical, deep learning, and hybrid models for AQI forecasting using daily AQI data from Jakarta during the 2023–2025 period. The dataset includes polutant parameters such as , , , , , and CO. The proposed models consist of Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM), and a hybrid ARIMA-LSTM model. The research methodology includes data preprocessing, normalization using Min-Max Scaling, sequence generation using the sliding window approach, model training, and evaluation using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The experimental results show that the CNN-LSTM model achieved the best forecasting performance with MAE, RMSE, and MAPE values of 4.54, 6.03, and 13.83%, respectively, followed closely by the LSTM model. Meanwhile, the ARIMA model produced the lowest performance, and the hybrid ARIMA-LSTM model did not outperform the standalone deep learning models. These findings indicate that deep learning approaches, particularly CNN-LSTM, are more effective in capturing the complex dynamics of urban air pollution data and have strong potential to support air quality forecasting and pollution control systems in Jakarta.
Cardiovascular diseases are the main reasons for death around the world at present, so early detection and intervention can be difficult. Review of Recent Applications of Machine Learning and Deep Learning in Cardiovascular Disease Prediction. Logistic regression, decision trees, random forests, support vector machines and gradient boosting have all been applied to the Cleveland and Kaggle cardiovascular datasets in previous studies. Based on research results, soft voting and stacking ensemble methods have been used to improve the prediction accuracy of a single classifier. Dense neural networks and hybrid Convolutional Neural Network - Long Short-Term Memory (CNN-LSTM) architectures are also deep learning models that have been researched and applied. Add to the above that SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) are popular methods for explaining Artificial Intelligence (AI) models. However, there are still many problems, such as an abundance of small public datasets, class imbalance, a lack of external validation, and poor clinical interpretability. The above are the problems of this paper, and some future directions for constructing a reliable cardiac disease prediction system with clinical applications are proposed.
Yi-Min Zhou· Applied and Computational En...· 0 citations
Chronic Kidney Disease (CKD) is a condition which is life-threatening,
progressive and if left untreated until an advanced stage of the disease, can
cause a lot of problems and cost. Noisy clinical data, feature redundancy,
inadequate representation learning, and poor generalization are common
limitations of existing machine learning and Deep Learning models. In order
to overcome these drawbacks, a hybrid deep ensemble model, namely
AutoCaps-LSTM-GBNet, is proposed to predict the CKD, which comprises
an Autoencoder (AE), Bidirectional Long Short-Term Memory (BiLSTM),
Capsule Network (CapsNet), and Gradient Boosted Deep Neural Network
(GB-DNN). The UCI data for CKD is initially pre-processed using KNN
imputation, Z-score normalization, SMOTE, and Recursive Feature
Elimination (RFE). The Autoencoder captures compact latent features and
BiLSTM and CapsNet capture complementary temporal and hierarchical
representations which are fused and classified using GB-DNN. An
experimental evaluation with repeated stratified 5-fold cross validation is
performed, which achieves an accuracy of 96.82 ± 1.94%, AUC of 0.983 ±
0.011, precision of 0.972, recall of 0.965, specificity of 0.968 and F1-score
of 0.968. The results show that this framework is well suited to making
accurate, reliable, and robust early CKD prediction which could be a good
candidate for computer-aided clinical decision-making support system
Unknown authors· European Journal of Prosthod...· 0 citations
Fine particulate matter (PM2.5) pollution in Jakarta reached an annual average of 37.3 µg/m³ in 2023, 7.4 times the WHO threshold, causing over 10,000 premature deaths annually. Accurate short-term prediction is essential for early-warning systems, yet two gaps remain common in decomposition-based deep learning literature: Variational Mode Decomposition (VMD) parameters are rarely selected via systematic sensitivity analysis, and prediction bias is rarely corrected explicitly. This study addresses both gaps using 32,144 PM2.5 observations from Jakarta (2015-2025). A grid search over 20 parameter combinations identified K=6, alpha=500 as optimal (reconstruction error 1.88%). Each of six IMFs was predicted independently using CNN1D, reconstructed additively, and bias-corrected using validation-set mean bias error. Decomposition, not model complexity, drove accuracy: a non-decomposed baseline reached only R²=0.4302, versus R²=0.9151 (RMSE=4.83 µg/m³) for the proposed framework. We further validated fixed- and adaptive-parameter variants (VMD, AVMD) across 5 independent runs. VMD-CNN1D achieved R²=0.9177±0.0173, RMSE=4.7346±0.5064, while AVMD-CNN1D (K=7 selected consistently) achieved R²=0.9238±0.0089, RMSE=4.5388±0.2649. Diebold-Mariano tests showed AVMD outperforming VMD in 4 of 5 runs (p<0.0001) with lower variance, though a paired t-test across runs was not significant (p=0.684). Ablation identified the lowest-frequency IMF as most critical, consistent with Jakarta's dry-season and land-fire pollution patterns, while residual analysis revealed heteroscedasticity as a limitation. These findings show that rigorous parameter selection, bias correction, and multi-run validation, not architectural complexity alone, make decomposition-based deep learning reliable for PM2.5 prediction, offering a reproducibility-aware baseline for a future Jakarta air-quality early-warning system.
Dwi Yuwono, A. A. Waskita, Tukiyat Tukiyat· JOURNAL OF APPLIED INFORMATI...· 0 citations
In the future, it is essential to establish a standardized validation framework, develop interpretable algorithms, integrate wearable non-invasive markers, and implement Bayesian racial modeling to promote early screening and personalized intervention, thereby revolutionizing the clinical prevention paradigm.
Dingnan Wu· Proceedings of the 3rd Inter...· 0 citations
Accurate diagnosis of cardiovascular disease is still a difficult problem, as clinical parameters show complex interactions and are continuously varying over time, making their prediction with conventional methods difficult. Current statistical and shallow learning methods often utilize handcrafted features, which are not capable of fully leveraging the vast information of patient data stored in multiple formats. We propose a hybrid deep learning framework in this study, which learns discriminative clinical representations and health patterns over time together with Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks. Clinical observations first are standardized to ensure uniformity in data and then passed through convolutional layers to create informative feature representations which are then improved by recurrent memory units to model temporal relationships between patient features. The developed framework has been applied using TensorFlow and tested on a benchmark dataset of cardiovascular diseases in the same experimental setup with conventional machine learning classifiers and individual deep learning models. The proposed architecture achieved 93.6% classification accuracy, 92.8% precision, 93.1% recall, 92.9% F1 score and ROC-AUC of 0.96 all of which showed consistent improvements over Logistic Regression, Decision Tree, Support Vector Machine, stand-alone CNN and stand-alone LSTM models.
M. M., A. S, K. B· ITM Web of Conferences· 0 citations
The early prediction of chronic illnesses such as diabetes, hypertension, and heart failure is vital for improving patient outcomes and reducing healthcare costs. However, Electronic Health Records (EHRs) present challenges due to their heterogeneous nature, containing a mix of structured (e.g., laboratory values, vital signs), unstructured (e.g., clinical notes), and temporal data (e.g., patient history). This paper proposes a novel deep ensemble learning framework that integrates Temporal Convolutional Networks (TCNs) and attention-guided Long Short-Term Memory (LSTM) networks for robust chronic disease prediction. The TCN module captures long-range temporal patterns and clinical trends across patient timelines, while the attention-enhanced LSTM module focuses on relevant time-steps and feature segments critical for diagnosis. The ensemble approach combines these deep representations to improve classification accuracy, reduce false positives, and provide resilience against data irregularities. Experiments conducted on benchmark multi-modal EHR datasets demonstrate the model’s superiority over traditional RNN, LSTM, and CNN-based methods. The proposed architecture not only enhances prediction performance but also contributes toward explainable and proactive healthcare decision-making.
T. Thamaraiselvan, K. Saravanan, S. Nithyanandam· Journal of Intelligent Decis...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
