The detection of gas emission levels is a crucial problem for ecology and human health. Hyperspectral image analysis offers many advantages over traditional gas detection systems with its detection capability from safe distances. Observing that the existing hyperspectral gas detection methods in the thermal range neglect the fact that the captured radiance the in longwave infrared (LWIR) spectrum is better modeled as a mixture of the radiance of background and target gases, we propose a deep learning-based hyperspectral gas detection method in this paper, which combines unmixing and classification. The proposed method first converts the radiance data to luminance-temperature data. Then, a 3D CNN and autoencoder-based network, which is specially designed for unmixing, is applied to the resulting data to acquire abundances and endmembers for each pixel. Finally, the detection is achieved by a three-layer fully connected network to detect the target gases at each pixel based on the extracted endmember spectra and abundance values. The superior performance of the proposed method with respect to the conventional hyperspectral gas detection methods using spectral angle mapper (SAM) and adaptive cosine estimator (ACE) is verified with LWIR hyperspectral images including Methane and Sulphur Dioxide gases. In addition, the ablation study with respect to different combinations of the proposed structure including direct classification and unmixing methods has revealed the contribution of the proposed system.

Hyperspectral image processing techniques, with their ability to provide information about the chemical compositions of materials, have great potential for pavement condition assessment. This study introduces a novel age-based pavement assessment method, employing an integrated algorithm with artificial neural network (ANN) and spectral angle mapping (SAM) on hyperspectral images. In the proposed method, the resulting ANN prediction outputs are used to make a new prediction along with the results from SAM scores. Tests are performed on hyperspectral images that have 360 spectral bands between 400 and 900 nm, collected by a specifically designed vehicular system for proximal image acquisition. The acquired images have eight classes, including three different pavement classes (good (5-year), medium (10-year), and poor (25-year)), yellow dye, white dye, soil, paving stone, and shadow. Several experiments are performed to evaluate the robustness of the followed methodology with limited learning data that include 5, 10, 25, and 50 samples per class, selected randomly from our independent spectral database. For a fair comparison, the individual ANN, SAM, support vector machine (SVM), and stacked auto-encoders (SAE) algorithms are also evaluated. The classification performances of individual ANN and SAM are significantly increased with their joint use, demonstrating a 1.2% to 21% classification accuracy improvement in relation to the training sample size. The study proves that the proposed approach is quite robust in cases wherein few training data are available, while SAE and standard ANN algorithms are more successful in cases wherein more learning data are present.

This article presents a new benchmark MRI dataset called the Gazi Brains Dataset 2020, containing MRI images of 100 patients, and introduces initial experimental results performed on this dataset in comparison with available brain MRI datasets. Furthermore, the dataset is analyzed using eight different deep learning models for high-grade glioma tumor prediction, classification, and detection tasks. Additionally, this study demonstrates the results of an explainable Artificial Intelligence (XAI) approach applied to the trained models. To demonstrate the utility of the proposed dataset, different deep learning models were applied to the problem, and these models were tested on various data and models applied for various tasks such as region of interest extraction, whole tumor segmentation, prediction, detection, and classification with accuracy, precision, recall, and F1-score. The experimental results indicate that the dataset is highly effective for multiple purposes, and the models reached significant results with successful F1-scores ranging between 93.2% and 96.4%. ROI and whole tumor segmentations were successfully performed and compared with seven algorithms with accuracies of 87.61% and 97.18%. The Grad-CAM model also demonstrated satisfactory accuracy across the tests that were conducted. Moreover, this study explores the application of XAI to the trained models, providing interpretability and insights into the decision-making processes. The findings signify that this dataset holds significant potential for various future research directions, including age estimation, gender detection, causal inference with XAI, and disease-related survival analysis.

Explosive detection is crucial for public safety and confidence. Among various solutions for this purpose, hyperspectral imaging (HSI) differs from its alternatives with its detection capability from standoff distances. However, the state of the art for such a technology is still significantly missing a complete technical and experimental framework for surveillance applications. In this paper, an end-to-end technical framework, which involves capturing, preprocessing, reflectance conversion, target detection, and performance evaluation stages, is proposed to reveal the potential of a ground-based hyperspectral image surveillance system for the detection of explosive traces. The proposed framework utilizes a shortwave infrared region (0.9-1.7μm), which covers the distinctive absorption characteristics of different explosives. Three classes of detection methods, namely index, signature, and learning-based methods are adapted to the proposed surveillance system. Their performances are compared over various experiments, which are specifically designed for granular and sprayed residues, fingerprint residues, and explosive traces on vehicles. The experiments reveal that the best method in terms of precision and recall performances is hybrid structure detector (HSD), which effectively combines signature-based detection with unmixing. While deep learning-based methods have also achieved satisfactory precision values, their low recall values for the moment have comparatively limited their usage for the high-risk cases. Although one of the main reasons for the current performances of deep learning methods is less data for learning, these performances for hyperspectral images can be increased with more data in the future as in other image applications.

Deep learning-based methods are accepted as a viable alternative to conventional statistical and geometrical methods for hyperspectral unmixing in recent years. These methods are however mainly based on linear mixture assumption on the hyperspectral data. The vast majority of presented algorithms process individual hyperspectral pixels while neglecting the spatial relationships between pixels. In order to address these two missing aspects, we propose a convolutional autoencoder-based hyperspectral unmixing method in this paper. The proposed structure incorporates the spatial neighbourhood relation with its convolutional layers in the first stage and possible non-linearities in the observed data with the included non-linear layer in the final stage. The experiments have first revealed that Adam optimizer have the best performance among different optimization methods for the proposed network. Second, the proposed method has indicated about 20–40% accuracy improvement in terms of mean squared error (MSE) metric compared to traditional hyperspectral abundance estimation methods. Third, the contribution of the non-linear layer is verified by comparing the proposed network with the conventional LMM-based autoencoder structure without the non-linear layer. Finally, the accuracy improvement for the proposed network with non-linear layer compared to the state-of-the-art deep learning-based methods using linear mixture assumption is evaluated in terms of MSE and reported as about 10% and 20% for synthetic and real data, respectively.

Polygenic risk scores are widely used for predicting genetic risk across complex diseases and traits, and several pre-trained models have been developed. Few approaches leverage these pre-trained polygenic risk scores to further refine predictive performance. Here, we present Adaptive Boosting of pre-trained Polygenic Risk Socres, a fine-tuning framework that refines pre-trained polygenic risk score models through adaptive variable selection and model boosting to identify additional predictive signals that may not be fully captured by the original models. Simulations show that our framework can identify signals orthogonal to pre-trained polygenic risk scores while controlling false discovery rates. Using UK Biobank data, we fine-tune pre-trained polygenic risk scores for binary diseases and continuous traits, and validate the results across three independent datasets: All of Us, eMERGE, and Penn Medicine Biobank. Real data analyses show that Adaptive Boosting of pre-trained Polygenic Risk Scores achieves statistically significant improvements in several scenarios while maintaining competitive performance in others.