Fusion of Raman and FTIR Spectroscopy Data Uncovers Physiological Changes Associated with Lung Cancer.

Due to the high mortality rate, more effective non-invasive diagnostic methods are still needed for lung cancer, the most common cause of cancer-related death worldwide. In this study, the integration of Raman and Fourier-transform infrared spectroscopy with advanced data-fusion techniques is investigated to improve the detection of lung cancer from human blood plasma samples. A high statistical significance was found for important protein-related oscillations, which are crucial for differentiating between lung cancer patients and healthy controls. The use of low-level data fusion and feature selection significantly improved model accuracy and emphasizes the importance of structural protein changes in cancer detection. Although other biomolecules such as carbohydrates and nucleic acids also contributed, proteins proved to be the decisive markers found using this technique. This research highlights the power of these combined spectroscopic methods to develop a non-invasive diagnostic tool for discriminating lung cancer from healthy state, with the potential to extend such studies to a variety of other diseases.

Comparison of deep-learning multimodality data fusion strategies in mandibular osteoradionecrosis NTCP modelling using clinical variables and radiation dose distribution volumes.

Objective.Normal tissue complication probability (NTCP) modelling is rapidly embracing deep learning (DL) methods, acknowledging the importance of spatial dose information. Finding effective ways to combine information from radiation dose distribution maps (dosiomics) and clinical data involves technical challenges and requires domain knowledge. We propose different multi-modality data fusion strategies to facilitate future DL-based NTCP studies.Approach.Early, joint and late DL multi-modality fusion strategies were compared using clinical and mandibular radiation dose distribution volumes. These were contrasted with single-modality models: a random forest trained on non-image data (clinical, demographic and dose-volume metrics) and a 3D DenseNet-40 trained on image data (mandibular dose distribution maps). The study involved a matched cohort of 92 osteoradionecrosis cases and 92 controls from a single institution.Main results.The late fusion model exhibited superior discrimination and calibration performance, while the join fusion achieved a more balanced distribution of the predicted probabilities. Discrimination performance did not significantly differ between strategies. Late fusion, though less technically complex, lacks crucial inter-modality interactions for NTCP modelling. In contrast, joint fusion, despite its complexity, resulted in a single network training process which included intra- and inter-modality interactions in its model parameter optimisation.Significance.This study is a pioneering effort in comparing different strategies for including image data into DL-based NTCP models in combination with lower dimensional data such as clinical variables. The discrimination performance of such multi-modality NTCP models and the choice of fusion strategy will depend on the distribution and quality of both types of data. Multiple data fusion strategies should be compared and reported in multi-modality NTCP modelling using DL.

A black tea quality testing method for scale production using CV and NIRS with TCN for spectral feature extraction.

To rigorously assess black tea quality in large-scale production, this study introduces a multi-modal fusion approach integrating computer vision (CV) with Near-Infrared Spectroscopy (NIRS). CV technology is first applied to evaluate the tea's appearance quality, while NIRS quantifies key chemical components, including tea polyphenols (TP), free amino acids (FAA), and caffeine (CAF). Additionally, different methods are employed to extract potential quality features from NIR spectra. The information are then fused, and a classifier is utilized to accurately identify tea quality. Results show that the Temporal Convolutional Network (TCN) fused model achieves a 98.2 % accuracy rate, surpassing both the Convolutional Neural Network (CNN) fused model and traditional methods. This study demonstrates that TCNs effectively extract spectral features and that data fusion significantly enhances tea quality testing, offering valuable insights for production optimization.

Integration of LC-HRMS and 1H NMR metabolomics data fusion approaches for classification of Amarone wine based on withering time and yeast strain.

Two untargeted metabolomics approaches (LC-HRMS and 1H NMR) were combined to classify Amarone wines based on grape withering time and yeast strain. The study employed a multi-omics data integration approach, combining unsupervised data exploration (MCIA) and supervised statistical analysis (sPLS-DA). The results revealed that the multi-omics pseudo-eigenvalue space highlighted a limited correlation between the datasets (RV-score = 16.4%), suggesting the complementarity of the assays. Furthermore, the sPLS-DA models correctly classified wine samples according to both withering time and yeast strains, providing a much broader characterization of wine metabolome with respect to what was obtained from the individual techniques. Significant variations were notably observed in the accumulation of amino acids, monosaccharides, and polyphenolic compounds throughout the withering process, with a lower error rate in sample classification (7.52%). In conclusion, this strategy demonstrated a high capability to integrate large omics datasets and identify key metabolites able to discriminate wine samples based on their characteristics.

Comparison of data fusion strategies for automated prostate lesion detection using mpMRI correlated with whole mount histology.

BACKGROUND: In this work, we compare input level, feature level and decision level data fusion techniques for automatic detection of clinically significant prostate lesions (csPCa). METHODS: Multiple deep learning CNN architectures were developed using the Unet as the baseline. The CNNs use both multiparametric MRI images (T2W, ADC, and High b-value) and quantitative clinical data (prostate specific antigen (PSA), PSA density (PSAD), prostate gland volume & gross tumor volume (GTV)), and only mp-MRI images (n = 118), as input. In addition, co-registered ground truth data from whole mount histopathology images (n = 22) were used as a test set for evaluation. RESULTS: The CNNs achieved for early/intermediate / late level fusion a precision of 0.41/0.51/0.61, recall value of 0.18/0.22/0.25, an average precision of 0.13 / 0.19 / 0.27, and F scores of 0.55/0.67/ 0.76. Dice Sorensen Coefficient (DSC) was used to evaluate the influence of combining mpMRI with parametric clinical data for the detection of csPCa. We compared the DSC between the predictions of CNN's trained with mpMRI and parametric clinical and the CNN's trained with only mpMRI images as input with the ground truth. We obtained a DSC of data 0.30/0.34/0.36 and 0.26/0.33/0.34 respectively. Additionally, we evaluated the influence of each mpMRI input channel for the task of csPCa detection and obtained a DSC of 0.14 / 0.25 / 0.28. CONCLUSION: The results show that the decision level fusion network performs better for the task of prostate lesion detection. Combining mpMRI data with quantitative clinical data does not show significant differences between these networks (p = 0.26/0.62/0.85). The results show that CNNs trained with all mpMRI data outperform CNNs with less input channels which is consistent with current clinical protocols where the same input is used for PI-RADS lesion scoring. TRIAL REGISTRATION: The trial was registered retrospectively at the German Register for Clinical Studies (DRKS) under proposal number Nr. 476/14 & 476/19.

Accuracy of manual and artificial intelligence-based superimposition of cone-beam computed tomography with digital scan data, utilizing an implant planning software: A randomized clinical study.

OBJECTIVES: To investigate the accuracy of conventional and automatic artificial intelligence (AI)-based registration of cone-beam computed tomography (CBCT) with intraoral scans and to evaluate the impact of user's experience, restoration artifact, number of missing teeth, and free-ended edentulous area. MATERIALS AND METHODS: Three initial registrations were performed for each of the 150 randomly selected patients, in an implant planning software: one from an experienced user, one from an inexperienced operator, and one from a randomly selected post-graduate student of implant dentistry. Six more registrations were performed for each dataset by the experienced clinician: implementing a manual or an automatic refinement, selecting 3 small or 3 large in-diameter surface areas and using multiple small or multiple large in-diameter surface areas. Finally, an automatic AI-driven registration was performed, using the AI tools that were integrated into the utilized implant planning software. The accuracy between each type of registration was measured using linear measurements between anatomical landmarks in metrology software. RESULTS: Fully automatic-based AI registration was not significantly different from the conventional methods tested for patients without restorations. In the presence of multiple restoration artifacts, user's experience was important for an accurate registration. Registrations' accuracy was affected by the number of free-ended edentulous areas, but not by the absolute number of missing teeth (p < .0083). CONCLUSIONS: In the absence of imaging artifacts, automated AI-based registration of CBCT data and model scan data can be as accurate as conventional superimposition methods. The number and size of selected superimposition areas should be individually chosen depending on each clinical situation.

Decision-level data fusion based on laser-induced breakdown and Raman spectroscopy: A study of bimodal spectroscopy for diagnosis of lung cancer at different stages.

Lung cancer staging is crucial for personalized treatment and improved prognosis. We propose a novel bimodal diagnostic approach that integrates LIBS and Raman technologies into a single platform, enabling comprehensive tissue elemental and molecular analysis. This strategy identifies critical staging elements and molecular marker signatures of lung tumors. LIBS detects concentration patterns of elemental lines including Mg (I), Mg (II), Ca (I), Ca (II), Fe (I), and Cu (II). Concurrently, Raman spectroscopy identifies changes in molecular content, such as phenylalanine (1033 cm-1), tyrosine (1174 cm-1), tryptophan (1207 cm-1), amide III (1267 cm-1), and proteins (1126 cm-1 and 1447 cm-1), among others. The bimodal information is fused using a decision-level Bayesian fusion model, significantly enhancing the performance of the convolutional neural network architecture in classification algorithms, with an accuracy of 99.17 %, sensitivity of 99.17 %, and specificity of 99.88 %. This study provides a powerful new tool for the accurate staging and diagnosis of lung tumors.

Multivariate data analysis on multisensor measurement for inline process monitoring of adenovirus production in HEK293 cells.

In the era of Biopharma 4.0, process digitalization fundamentally requires accurate and timely monitoring of critical process parameters (CPPs) and quality attributes. Bioreactor systems are equipped with a variety of sensors to ensure process robustness and product quality. However, during the biphasic production of viral vectors or replication-competent viruses for gene and cell therapies and vaccination, current monitoring techniques relying on a single working sensor can be affected by the physiological state change of the cells due to infection/transduction/transfection step required to initiate production. To address this limitation, a multisensor (MS) monitoring system, which includes dual-wavelength fluorescence spectroscopy, dielectric signals, and a set of CPPs, such as oxygen uptake rate and pH control outputs, was employed to monitor the upstream process of adenovirus production in HEK293 cells in bioreactor. This system successfully identified characteristic responses to infection by comparing variations in these signals, and the correlation between signals and target critical variables was analyzed mechanistically and statistically. The predictive performance of several target CPPs using different multivariate data analysis (MVDA) methods on data from a single sensor/source or fused from multiple sensors were compared. An MS regression model can accurately predict viable cell density with a relative root mean squared error (rRMSE) as low as 8.3% regardless of the changes occurring over the infection phase. This is a significant improvement over the 12% rRMSE achieved with models based on a single source. The MS models also provide the best predictions for glucose, glutamine, lactate, and ammonium. These results demonstrate the potential of using MVDA on MS systems as a real-time monitoring approach for biphasic bioproduction processes. Yet, models based solely on the multiplicity and timing of infection outperformed both single-sensor and MS models, emphasizing the need for a deeper mechanistic understanding in virus production prediction.

PheSeq, a Bayesian deep learning model to enhance and interpret the gene-disease association studies.

Despite the abundance of genotype-phenotype association studies, the resulting association outcomes often lack robustness and interpretations. To address these challenges, we introduce PheSeq, a Bayesian deep learning model that enhances and interprets association studies through the integration and perception of phenotype descriptions. By implementing the PheSeq model in three case studies on Alzheimer's disease, breast cancer, and lung cancer, we identify 1024 priority genes for Alzheimer's disease and 818 and 566 genes for breast cancer and lung cancer, respectively. Benefiting from data fusion, these findings represent moderate positive rates, high recall rates, and interpretation in gene-disease association studies.

A muti-modal feature fusion method based on deep learning for predicting immunotherapy response.

Immune checkpoint therapy (ICT) has greatly improved the survival of cancer patients in the past few years, but only a small number of patients respond to ICT. To predict ICT response, we developed a multi-modal feature fusion model based on deep learning (MFMDL). This model utilizes graph neural networks to map gene-gene relationships in gene networks to low dimensional vector spaces, and then fuses biological pathway features and immune cell infiltration features to make robust predictions of ICT. We used five datasets to validate the predictive performance of the MFMDL. These five datasets span multiple types of cancer, including melanoma, lung cancer, and gastric cancer. We found that the prediction performance of multi-modal feature fusion model based on deep learning is superior to other traditional ICT biomarkers, such as ICT targets or tumor microenvironment-associated markers. In addition, we also conducted ablation experiments to demonstrate the necessity of fusing different modal features, which can improve the prediction accuracy of the model.

Interpretable deep learning methods for multiview learning.

BACKGROUND: Technological advances have enabled the generation of unique and complementary types of data or views (e.g. genomics, proteomics, metabolomics) and opened up a new era in multiview learning research with the potential to lead to new biomedical discoveries. RESULTS: We propose iDeepViewLearn (Interpretable Deep Learning Method for Multiview Learning) to learn nonlinear relationships in data from multiple views while achieving feature selection. iDeepViewLearn combines deep learning flexibility with the statistical benefits of data and knowledge-driven feature selection, giving interpretable results. Deep neural networks are used to learn view-independent low-dimensional embedding through an optimization problem that minimizes the difference between observed and reconstructed data, while imposing a regularization penalty on the reconstructed data. The normalized Laplacian of a graph is used to model bilateral relationships between variables in each view, therefore, encouraging selection of related variables. iDeepViewLearn is tested on simulated and three real-world data for classification, clustering, and reconstruction tasks. For the classification tasks, iDeepViewLearn had competitive classification results with state-of-the-art methods in various settings. For the clustering task, we detected molecular clusters that differed in their 10-year survival rates for breast cancer. For the reconstruction task, we were able to reconstruct handwritten images using a few pixels while achieving competitive classification accuracy. The results of our real data application and simulations with small to moderate sample sizes suggest that iDeepViewLearn may be a useful method for small-sample-size problems compared to other deep learning methods for multiview learning. CONCLUSION: iDeepViewLearn is an innovative deep learning model capable of capturing nonlinear relationships between data from multiple views while achieving feature selection. It is fully open source and is freely available at https://github.com/lasandrall/iDeepViewLearn .

Machine Learning Methods for Gene Selection in Uveal Melanoma.

Uveal melanoma (UM) is the most common primary intraocular malignancy with a limited five-year survival for metastatic patients. Limited therapeutic treatments are currently available for metastatic disease, even if the genomics of this tumor has been deeply studied using next-generation sequencing (NGS) and functional experiments. The profound knowledge of the molecular features that characterize this tumor has not led to the development of efficacious therapies, and the survival of metastatic patients has not changed for decades. Several bioinformatics methods have been applied to mine NGS tumor data in order to unveil tumor biology and detect possible molecular targets for new therapies. Each application can be single domain based while others are more focused on data integration from multiple genomics domains (as gene expression and methylation data). Examples of single domain approaches include differentially expressed gene (DEG) analysis on gene expression data with statistical methods such as SAM (significance analysis of microarray) or gene prioritization with complex algorithms such as deep learning. Data fusion or integration methods merge multiple domains of information to define new clusters of patients or to detect relevant genes, according to multiple NGS data. In this work, we compare different strategies to detect relevant genes for metastatic disease prediction in the TCGA uveal melanoma (UVM) dataset. Detected targets are validated with multi-gene score analysis on a larger UM microarray dataset.

Steel Automotive Wheel Rims-Data Fusion for the Precise Identification of the Technical Condition and Indication of the Approaching End of Service Life.

Steel automotive wheel rims are subject to wear and tear, down to the end of their service life. Manufacturers use standard destructive tests to determine the probable lifetime of the car wheel rim. With this approach, to predict the remaining use time, it is necessary to know the initial parameters of the wheel rim, actual mileage, and its use characteristics, which is difficult information to obtain in the real world. Moreover, this work shows that a vehicle's technical condition can affect the rim's remaining service time. This work describes a new method of precise binary identification of the technical condition of steel car wheel rims using the dispersion of damping factors which result from experimental modal analysis. This work also proposes a new method of indicating the approaching end of wheel rim service life with limited parameters: run-out, average of damping factors, and dispersion of damping factors. The proposed procedure requires two sequential examinations of the rim in standard periods related to the average annual mileage of the vehicle. On this basis, it is possible to indicate the approaching end of the life of the steel rims about 10,000 km in advance.

Transcriptome analysis reveals involvement of thiopurine S-methyltransferase in oxidation-reduction processes.

Thiopurine S-methyltransferase (TPMT) is an important enzyme involved in the deactivation of thiopurines and represents a major determinant of thiopurine-related toxicities. Despite its well-known importance in thiopurine metabolism, the understanding of its endogenous role is lacking. In the present study, we aimed to gain insight into the molecular processes involving TPMT by applying a data fusion approach to analyze whole-genome expression data. The RNA profiling was done on whole blood samples from 1017 adult male and female donors to the Estonian biobank using Illumina HTv3 arrays. Our results suggest that TPMT is closely related to genes involved in oxidoreductive processes. The in vitro experiments on different cell models confirmed that TPMT influences redox capacity of the cell by altering S-adenosylmethionine (SAM) consumption and consequently glutathione (GSH) synthesis. Furthermore, by comparing gene networks of subgroups of individuals, we identified genes, which could have a role in regulating TPMT activity. The biological relevance of identified genes and pathways will have to be further evaluated in molecular studies.

Big data and artificial intelligence in cancer research.

The field of oncology has witnessed an extraordinary surge in the application of big data and artificial intelligence (AI). AI development has made multiscale and multimodal data fusion and analysis possible. A new era of extracting information from complex big data is rapidly evolving. However, challenges related to efficient data curation, in-depth analysis, and utilization remain. We provide a comprehensive overview of the current state of the art in big data and computational analysis, highlighting key applications, challenges, and future opportunities in cancer research. By sketching the current landscape, we seek to foster a deeper understanding and facilitate the advancement of big data utilization in oncology, call for interdisciplinary collaborations, ultimately contributing to improved patient outcomes and a profound understanding of cancer.

Effects of Indoor Dust Exposure on Lung Cells: Association of Chemical Composition with Phenotypic and Lipid Changes in a 3D Lung Cancer Cell Model.

Indoor dust is a key contributor to the global human exposome in urban areas since the population develops most of its activities in private and public buildings. To gain insight into the health risks associated with this chronic exposure, it is necessary to characterize the chemical composition of dust and understand its biological impacts using reliable physiological models. The present study investigated the biological effects of chemically characterized indoor dust extracts using three-dimensional (3D) lung cancer cell cultures combining phenotypic and lipidomic analyses. Apart from the assessment of cell viability, reactive oxygen species (ROS) induction, and interleukin-8 release, lipidomics was applied to capture the main lipid changes induced as a cellular response to the extracted dust compounds. The application of chemometric tools enabled the finding of associations between chemical compounds present in dust and lipidic and phenotypic profiles in the cells. This study contributes to a better understanding of the toxicity mechanisms associated with exposure to chemical pollutants present in indoor dust.

Comparing 1.5 T and 3.0 T MR data for 3D visualization of neurovascular relationships in the posterior fossa.

BACKGROUND: Neurovascular relationships in the posterior fossa are more frequently investigated due to the increasing availability of 3.0 Tesla MRI. For an assessment with 3D visualization, no systematic analyzes are available so far and the question arises as to whether 3.0 Tesla MRI should be given preference over 1.5 Tesla MRI. METHODS: In a prospective study, a series of 25 patients each underwent MRI investigations with 3D-CISS and 3D-TOF at 1.5 and 3.0 Tesla. For both field strengths separately, blood vessel information from the TOF data was fused into the CISS data after segmentation and registration. Four visualizations were created for each field strength, with and without optimization before and after fusion, which were evaluated with a rating system and verified with the intraoperative situation. RESULTS: When only CISS data was used, nerves and vessels were better visualized at 1.5 Tesla. After fusion, flow and pulsation artifacts were reduced in both cases, missing vessel sections were supplemented at 3.0 Tesla and 3D visualization at 1.5 and 3.0 Tesla led to anatomically comparable results. By subsequent manual correction, the remaining artifacts were further eliminated, with the 3D visualization being significantly better at 3.0 Tesla, since the higher field strength led to sharper contours of small vessel and nerve structures. CONCLUSION: 3D visualizations at 1.5 Tesla are sufficiently detailed for planning microvascular decompression and can be used without restriction. Fusion further improves the quality of 3D visualization at 3.0 Tesla and enables an even more accurate delineation of cranial nerves and vessels.

Developing reliable classification of dual IDO1/TDO inhibitors using data fusion and majority voting.

Indoleamine 2,3-dioxygenase 1 (IDO1) and tryptophan 2,3-dioxygenase (TDO) are promising dual-targeting inhibitors in cancer and neurodegenerative diseases treatment. Data fusion of receptor-based and ligand-based information of dual IDO1/TDO inhibitors were employed for active/inactive classification performance. A reliable decision making procedure was used here to identify active/inactive dual IDO1/TDO inhibitors using majority voting method and pools of individual classifications instead of individual models. All classification models were validated using prediction set, cross-validation and y-scrambling tests. The classification outcomes indicate that the sensitivity, specificity, precision, accuracy, G-mean and F1 score values increases up to ∼90% using data fusion and majority voting method. Compare to individual classification models with a single prediction point, the majority voting method has more reliable results due to the integration of the pool of individual classification models. This classification strategy may lead to more reliable identification of active/inactive dual-targeting inhibitors in cancer immunotherapy.Communicated by Ramaswamy H. Sarma.

Full-Perception Robotic Surgery Environment with Anti-Occlusion Global-Local Joint Positioning.

The robotic surgery environment represents a typical scenario of human-robot cooperation. In such a scenario, individuals, robots, and medical devices move relative to each other, leading to unforeseen mutual occlusion. Traditional methods use binocular OTS to focus on the local surgical site, without considering the integrity of the scene, and the work space is also restricted. To address this challenge, we propose the concept of a fully perception robotic surgery environment and build a global-local joint positioning framework. Furthermore, based on data characteristics, an improved Kalman filter method is proposed to improve positioning accuracy. Finally, drawing from the view margin model, we design a method to evaluate positioning accuracy in a dynamic occlusion environment. The experimental results demonstrate that our method yields better positioning results than classical filtering methods.

Orientation-Independent Human Activity Recognition Using Complementary Radio Frequency Sensing.

RF sensing offers an unobtrusive, user-friendly, and privacy-preserving method for detecting accidental falls and recognizing human activities. Contemporary RF-based HAR systems generally employ a single monostatic radar to recognize human activities. However, a single monostatic radar cannot detect the motion of a target, e.g., a moving person, orthogonal to the boresight axis of the radar. Owing to this inherent physical limitation, a single monostatic radar fails to efficiently recognize orientation-independent human activities. In this work, we present a complementary RF sensing approach that overcomes the limitation of existing single monostatic radar-based HAR systems to robustly recognize orientation-independent human activities and falls. Our approach used a distributed mmWave MIMO radar system that was set up as two separate monostatic radars placed orthogonal to each other in an indoor environment. These two radars illuminated the moving person from two different aspect angles and consequently produced two time-variant micro-Doppler signatures. We first computed the mean Doppler shifts (MDSs) from the micro-Doppler signatures and then extracted statistical and time- and frequency-domain features. We adopted feature-level fusion techniques to fuse the extracted features and a support vector machine to classify orientation-independent human activities. To evaluate our approach, we used an orientation-independent human activity dataset, which was collected from six volunteers. The dataset consisted of more than 1350 activity trials of five different activities that were performed in different orientations. The proposed complementary RF sensing approach achieved an overall classification accuracy ranging from 98.31 to 98.54%. It overcame the inherent limitations of a conventional single monostatic radar-based HAR and outperformed it by 6%.