Background: Lung and colon cancers are among the most prevalent and lethal malignancies worldwide, underscoring the urgent need for advanced diagnostic methodologies. This study aims to develop a hybrid deep learning and machine learning framework for the classification of Colon Adenocarcinoma, Colon Benign Tissue, Lung Adenocarcinoma, Lung Benign Tissue, and Lung Squamous Cell Carcinoma from histopathological images. Methods: Current approaches primarily rely on the LC25000 dataset, which, due to image augmentation, lacks the generalizability required for real-time clinical applications. To address this, Contrast Limited Adaptive Histogram Equalization (CLAHE) was applied to enhance image quality, and 1000 new images from the National Cancer Institute GDC Data Portal were introduced into the Colon Adenocarcinoma, Lung Adenocarcinoma, and Lung Squamous Cell Carcinoma classes, replacing augmented images to increase dataset diversity. A hybrid feature extraction model combining MobileNetV2 and EfficientNetB3 was optimized using the Grey Wolf Optimizer (GWO), resulting in the Lung and Colon histopathological classification technique (MEGWO-LCCHC). Cross-validation and hyperparameter tuning with Optuna were performed on various machine learning models, including XGBoost, LightGBM, and CatBoost. Results: The MEGWO-LCCHC technique achieved high classification accuracy, with the lightweight DNN model reaching 94.8%, LightGBM at 93.9%, XGBoost at 93.5%, and CatBoost at 93.3% on the test set. Conclusions: The findings suggest that our approach enhances classification performance and offers improved generalizability for real-world clinical applications. The proposed MEGWO-LCCHC framework shows promise as a robust tool in cancer diagnostics, advancing the application of AI in oncology.

Volatile organic compounds (VOCs) in exhaled human breath serve as pivotal biomarkers for disease identification and medical diagnostics. In the context of diabetes mellitus, the noninvasive detection of acetone, a primary biomarker using electronic noses (e-noses), has gained significant attention. However, employing e-noses requires pre-trained algorithms for precise diabetes detection, often requiring a computer with a programming environment to classify newly acquired data. This study focuses on the development of an embedded system integrating Tiny Machine Learning (TinyML) and an e-nose equipped with Metal Oxide Semiconductor (MOS) sensors for real-time diabetes detection. The study encompassed 44 individuals, comprising 22 healthy individuals and 22 diagnosed with various types of diabetes mellitus. Test results highlight the XGBoost Machine Learning algorithm’s achievement of 95% detection accuracy. Additionally, the integration of deep learning algorithms, particularly deep neural networks (DNNs) and one-dimensional convolutional neural network (1D-CNN), yielded a detection efficacy of 94.44%. These outcomes underscore the potency of combining e-noses with TinyML in embedded systems, offering a noninvasive approach for diabetes mellitus detection.

Diabetes mellitus, a chronic condition affecting millions worldwide, necessitates continuous monitoring of blood glucose level (BGL). The increasing prevalence of diabetes has driven the development of non-invasive methods, such as electronic noses (e-noses), for analyzing exhaled breath and detecting biomarkers in volatile organic compounds (VOCs). Effective machine learning models require extensive patient data to ensure accurate BGL predictions, but previous studies have been limited by small sample sizes. This study addresses this limitation by employing conditional generative adversarial networks (CTGAN) to generate synthetic data from real-world tests involving 29 healthy and 29 diabetic participants, resulting in over 14,000 new synthetic samples. These data were used to validate machine learning models for diabetes detection and BGL prediction, integrated into a Tiny Machine Learning (TinyML) e-nose system for real-time analysis. The proposed models achieved an 86% accuracy in BGL identification using LightGBM (Light Gradient Boosting Machine) and a 94.14% accuracy in diabetes detection using Random Forest. These results demonstrate the efficacy of enhancing machine learning models with both real and synthetic data, particularly in non-invasive systems integrating e-noses with TinyML. This study signifies a major advancement in non-invasive diabetes monitoring, underscoring the transformative potential of TinyML-powered e-nose systems in healthcare applications.

Background/Objectives: The increasing prevalence of diabetes underscores the urgent need for non-invasive, rapid, and cost-effective diagnostic alternatives. This study presents a breath-based multiclass diabetes classification system leveraging only three gas sensors (CO, alcohol, and acetone) to analyze exhaled breath composition. Methods: Breath samples were collected from 58 participants (22 healthy, 7 prediabetic, and 29 diabetic), with blood glucose levels serving as the reference metric. To enhance classification performance, we introduced a novel biomarker, the alcohol-to-acetone ratio, through a feature engineering approach. Class imbalance was addressed using the Synthetic Minority Over-Sampling Technique (SMOTE), ensuring a balanced dataset for model training. A nested cross-validation framework with 3 outer and 3 inner folds was implemented. Multiple machine learning classifiers were evaluated, with Random Forest and Gradient Boosting emerging as the top-performing models. Results: An ensemble combining both yielded the highest overall performance, achieving an average accuracy of 98.86%, precision of 99.07%, recall of 98.81% and F1 score of 98.87%. These findings highlight the potential of gas sensor-based breath analysis as a highly accurate, scalable, and non-invasive method for diabetes screening. Conclusions: The proposed system offers a promising alternative to blood-based diagnostic approaches, paving the way for real-world applications in point-of-care diagnostics and continuous health monitoring.

Lung and colon cancers are among the most common and lethal types of cancer. It is possible for individuals to develop both cancers simultaneously, as risk factors such as smoking, which is a leading cause of lung cancer, can also contribute to poor dietary habits, thereby increasing the risk of colon cancer. Traditionally, the diagnosis of these cancers relies on biopsies and subsequent laboratory analysis. This study proposes an Inception-ResNetV2-based model for accurately classifying lung and colon cancer using histopathological images. The dataset analyzed consists of 25,000 images categorized into five cancer types: Colon Adenocarcinoma, Colon Benign Tissue, Lung Adenocarcinoma, Lung Benign Tissue, and Lung Squamous Cell Carcinoma. The developed approach achieved an accuracy of 99.15%. The model's performance is further enhanced through the use of Local Binary Patterns (LBP), improving both accuracy and computational efficiency. Additionally, the explainable AI method SHAP is employed to demonstrate the contribution of each feature to the predictive analysis