Interpretable Deep Learning Model Research Articles

An interpretable deep learning model predicts RNA–small molecule binding sites

The intricate interplay between RNA molecules and small ligands plays a pivotal role in regulating biological processes, underscoring the necessity for accurate prediction models of RNA–small molecule binding sites. In this study, we develop CapBind, a novel interpretive deep learning framework utilizing a Capsule Network architecture, to precisely identify RNA–small molecule binding sites. This framework stands out for its multifaceted approach, which leverages both the sequence and structural information of RNA as input features, allowing for a thorough understanding of the complex interactions between RNA and small molecules. Leveraging multi-source bioinformatics data, including the distribution of RNA secondary structures, our model characterizes sequence and structural features with unparalleled depth. Through a unified alignment of these multi-source feature representations, we achieve dimensional homogeneity in the potential feature maps for both types of biological data. The crux of our methodology lies in the deployment of a multi-level Capsule Network, engineered to identify the nuanced interactions of RNA sequences with small molecules. The essence of our approach resides in the implementation of a multi-layered Capsule Network. Utilizing a sophisticated routing mechanism, this network demonstrates unparalleled efficacy in learning and identifying the intricate interactions between RNA sequences and small molecules. This mechanism enables dynamic adjustment of information flow across multiple levels, thereby capturing subtle features and patterns that might elude conventional neural network architectures. The experimental results demonstrate the superior performance of our model, particularly highlighting its progress in evaluation metrics compared to current state-of-the-art models. Case studies further validate the model’s precision in identifying binding sites, offering potential insights into gene expression pathways and novel therapeutic avenues for neurodegenerative diseases. Moreover, an extensive interpretability analysis of our silicon model reveals the internal interaction mechanisms between specific RNA regions and small molecules, contributing valuable perspectives for drug design and molecular biology research. These findings underscore the model’s predictive accuracy and biological applicability, establishing our approach as a significant leap forward in the computational prediction of RNA–small molecule binding sites.

Prediction of anticancer drug sensitivity using an interpretable model guided by deep learning

BackgroundThe prediction of drug sensitivity plays a crucial role in improving the therapeutic effect of drugs. However, testing the effectiveness of drugs is challenging due to the complex mechanism of drug reactions and the lack of interpretability in most machine learning and deep learning methods. Therefore, it is imperative to establish an interpretable model that receives various cell line and drug feature data to learn drug response mechanisms and achieve stable predictions between available datasets.ResultsThis study proposes a new and interpretable deep learning model, DrugGene, which integrates gene expression, gene mutation, gene copy number variation of cancer cells, and chemical characteristics of anticancer drugs to predict their sensitivity. This model comprises two different branches of neural networks, where the first involves a hierarchical structure of biological subsystems that uses the biological processes of human cells to form a visual neural network (VNN) and an interpretable deep neural network for human cancer cells. DrugGene receives genotype input from the cell line and detects changes in the subsystem states. We also employ a traditional artificial neural network (ANN) to capture the chemical structural features of drugs. DrugGene generates final drug response predictions by combining VNN and ANN and integrating their outputs into a fully connected layer. The experimental results using drug sensitivity data extracted from the Cancer Drug Sensitivity Genome Database and the Cancer Treatment Response Portal v2 reveal that the proposed model is better than existing prediction methods. Therefore, our model achieves higher accuracy, learns the reaction mechanisms between anticancer drugs and cell lines from various features, and interprets the model’s predicted results.ConclusionsOur method utilizes biological pathways to construct neural networks, which can use genotypes to monitor changes in the state of network subsystems, thereby interpreting the prediction results in the model and achieving satisfactory prediction accuracy. This will help explore new directions in cancer treatment. More available code resources can be downloaded for free from GitHub (https://github.com/pangweixiong/DrugGene).

Pseudo-class part prototype networks for interpretable breast cancer classification

Interpretability in machine learning has become increasingly important as machine learning is being used in more and more applications, including those with high-stakes consequences such as healthcare where Interpretability has been regarded as a key to the successful adoption of machine learning models. However, using confounding/irrelevant information in making predictions by deep learning models, even the interpretable ones, poses critical challenges to their clinical acceptance. That has recently drawn researchers’ attention to issues beyond the mere interpretation of deep learning models. In this paper, we first investigate application of an inherently interpretable prototype-based architecture, known as ProtoPNet, for breast cancer classification in digital pathology and highlight its shortcomings in this application. Then, we propose a new method that uses more medically relevant information and makes more accurate and interpretable predictions. Our method leverages the clustering concept and implicitly increases the number of classes in the training dataset. The proposed method learns more relevant prototypes without any pixel-level annotated data. To have a more holistic assessment, in addition to classification accuracy, we define a new metric for assessing the degree of interpretability based on the comments of a group of skilled pathologists. Experimental results on the BreakHis dataset show that the proposed method effectively improves the classification accuracy and interpretability by respectively 8%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$8 \\%$$\\end{document} and 18%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$18 \\%$$\\end{document}. Therefore, the proposed method can be seen as a step toward implementing interpretable deep learning models for the detection of breast cancer using histopathology images.

Toward better atmospheric polycyclic aromatic hydrocarbons pollution control in the Northern Hemisphere: Process analysis based on interpretable deep learning models

The accurate prediction of atmospheric polycyclic aromatic hydrocarbons (PAHs) concentrations in the Arctic is crucial in guiding the development of pollution control measures for atmospheric PAHs. However, the complex formation process and strong time-dependence of PAHs in the Arctic increase the difficulty of prediction. Herein, deep learning (DL) models were developed to predict monthly air concentrations of 16 priority PAHs at Alert monitoring station by incorporating the inherent periodicity of PAH concentrations and ancillary data including PAH emissions, meteorological information, fire emissions, and sea ice area. The prediction performance of long short-term memory (LSTM), gated recurrent unit (GRU) and bidirectional LSTM (Bi LSTM) algorithms were compared by multiple model evaluation metrics. The results showed that Bi LSTM models outperformed LSTM and GRU in predicting PAHs concentrations because it exhibited higher accuracy in terms of average R2ext (0.09) and errors (average MAPEext = 1.45). The DL interpretation based on SHAP values suggested that the main drivers of PAHs concentrations were PAHs emissions and sea ice area, with a more prominent contribution from meteorological conditions. The interpretable deep learning approach provides a potential shortcut for predicting time-delayed PAHs concentrations in the Arctic, promoting the management and control of global PAHs pollution.

Interpretable deep learning models for the inference and classification of LHC data

The Shower Deconstruction methodology is pivotal in distinguishing signal and background jets, leveraging the detailed information from perturbative parton showers. Rooted in the Neyman-Pearson lemma, this method is theoretically designed to differentiate between signal and background processes optimally in high-energy physics experiments. A key challenge, however, arises from the combinatorial growth associated with increasing jet constituents, which hampers its computational feasibility. We address this by demonstrating that the likelihood derived from comparing the most probable signal and background shower histories is equally effective for discrimination as the conventional approach of summing over all potential histories in top quark versus Quantum Chromodynamics (QCD) scenarios. We propose a novel approach by conceptualising the identification of the most probable shower history as a Markov Decision Process (MDP). Utilising a sophisticated modular point-transformer architecture, our method efficiently learns the optimal policy for this task. The developed neural agent excels in constructing the most likely shower history and demonstrates robust generalisation capabilities on unencountered test data. Remarkably, our approach mitigates the complexity inherent in the inference process, achieving a linear scaling relationship with the number of jet constituents. This offers a computationally viable and theoretically sound method for signal-background differentiation, paving the way for more effective data analysis in particle physics.

On the evaluation of deep learning interpretability methods for medical images under the scope of faithfulness

Background and Objective:Evaluating the interpretability of Deep Learning models is crucial for building trust and gaining insights into their decision-making processes. In this work, we employ class activation map based attribution methods in a setting where only High-Resolution Class Activation Mapping (HiResCAM) is known to produce faithful explanations. The objective is to evaluate the quality of the attribution maps using quantitative metrics and investigate whether faithfulness aligns with the metrics results. Methods:We fine-tune pre-trained deep learning architectures over four medical image datasets in order to calculate attribution maps. The maps are evaluated on a threefold metrics basis utilizing well-established evaluation scores. Results:Our experimental findings suggest that the Area Over Perturbation Curve (AOPC) and Max-Sensitivity scores favor the HiResCAM maps. On the other hand, the Heatmap Assisted Accuracy Score (HAAS) does not provide insights to our comparison as it evaluates almost all maps as inaccurate. To this purpose we further compare our calculated values against values obtained over a diverse group of models which are trained on non-medical benchmark datasets, to eventually achieve more responsive results. Conclusion:This study develops a series of experiments to discuss the connection between faithfulness and quantitative metrics over medical attribution maps. HiResCAM preserves the gradient effect on a pixel level ultimately producing high-resolution, informative and resilient mappings. In turn, this is depicted in the results of AOPC and Max-Sensitivity metrics, successfully identifying the faithful algorithm. In regards to HAAS, our experiments yield that it is sensitive over complex medical patterns, commonly characterized by strong color dependency and multiple attention areas.

Variational Attention-Based Interpretable Transformer Network for Rotary Machine Fault Diagnosis.

Deep learning technology provides a promising approach for rotary machine fault diagnosis (RMFD), where vibration signals are commonly utilized as input of a deep network model to reveal the internal state of machinery. However, most existing methods fail to mine association relationships within signals. Unlike deep neural networks, transformer networks are capable of capturing association relationships through the global self-attention mechanism to enhance feature representations from vibration signals. Despite this, transformer networks cannot explicitly establish the causal association between signal patterns and fault types, resulting in poor interpretability. To tackle these problems, an interpretable deep learning model named the variational attention-based transformer network (VATN) is proposed for RMFD. VATN is improved from transformer encoder to mine the association relationships within signals. To embed the prior knowledge of the fault type, which can be recognized based on several key features of vibration signals, a sparse constraint is designed for attention weights. Variational inference is employed to force attention weights to samples from Dirichlet distributions, and Laplace approximation is applied to realize reparameterization. Finally, two experimental studies conducted on bevel gear and bearing datasets demonstrate the effectiveness of VATN to other comparison methods, and the heat map of attention weights illustrates the causal association between fault types and signal patterns.

Deep learning the cis-regulatory code for gene expression in selected model plants

Elucidating the relationship between non-coding regulatory element sequences and gene expression is crucial for understanding gene regulation and genetic variation. We explored this link with the training of interpretable deep learning models predicting gene expression profiles from gene flanking regions of the plant species Arabidopsis thaliana, Solanum lycopersicum, Sorghum bicolor, and Zea mays. With over 80% accuracy, our models enabled predictive feature selection, highlighting e.g. the significant role of UTR regions in determining gene expression levels. The models demonstrated remarkable cross-species performance, effectively identifying both conserved and species-specific regulatory sequence features and their predictive power for gene expression. We illustrated the application of our approach by revealing causal links between genetic variation and gene expression changes across fourteen tomato genomes. Lastly, our models efficiently predicted genotype-specific expression of key functional gene groups, exemplified by underscoring known phenotypic and metabolic differences between Solanum lycopersicum and its wild, drought-resistant relative, Solanum pennellii.

Using an interpretable deep learning model for the prediction of riverine suspended sediment load.

The prediction of suspended sediment load (SSL) within riverine systems is critical to understanding the watershed's hydrology. Therefore, the novelty of our research is developing an interpretable (explainable) model based on deep learning (DL) and Shapley Additive ExPlanations (SHAP) interpretation technique for prediction of SSL in the riverine systems. This paper investigates the abilities of four DL models, including dense deep neural networks (DDNN), long short-term memory (LSTM), gated recurrent unit (GRU), and simple recurrent neural network (RNN) models for the prediction of daily SSL using river discharge and rainfall data at a daily time scale in the Taleghan River watershed, northwestern Tehran, Iran. The performance of models was evaluated by using several quantitative and graphical criteria. The effect of parameter settings on the performance of deep models on SSL prediction was also investigated. The optimal optimization algorithms, maximum iteration (MI), and batch size (BC) were obtained for modeling daily SSL, and structure of the model impact on prediction remarkably. The comparison of prediction accuracy of the models illustrated that DDNN (with R2 = 0.96, RMSE = 333.46) outperformed LSTM (R2 = 0.75, RMSE = 786.20), GRU (R2 = 0.73, RMSE = 825.67), and simple RNN (R2 = 0.78, RMSE = 741.45). Furthermore, the Taylor diagram confirmed that DDNN has the highest performance among other models. Interpretation techniques can address the black-box nature of models, and here, SHAP was applied to develop an interpretable DL model to interpret of DL model's output. The results of SHAP showed that river discharge has the strongest impact on the model's output in estimating SSL. Overall, we conclude that DL models have great potential in watersheds to predict SSL. Therefore, different interpretation techniques as tools to interpret DL model's output (DL model is as black-box model) are recommended in future research.

Development of an Interpretable Deep Learning Model for Pathological Tumor Response Assessment After Neoadjuvant Therapy.

Neoadjuvant therapy followed by surgery has become the standard of care for locally advanced esophageal squamous cell carcinoma (ESCC) and accurate pathological response assessment is critical to assess the therapeutic efficacy. However, it can be laborious and inconsistency between different observers may occur. Hence, we aim to develop an interpretable deep-learning model for efficient pathological response assessment following neoadjuvant therapy in ESCC. This retrospective study analyzed 337 ESCC resection specimens from 2020-2021 at the Pudong-Branch (Cohort 1) and 114 from 2021-2022 at the Puxi-Branch (External Cohort 2) of Fudan University Shanghai Cancer Center. Whole slide images (WSIs) from these two cohorts were generated using different scanning machines to test the ability of the model in handling color variations. Four pathologists independently assessed the pathological response. The senior pathologists annotated tumor beds and residual tumor percentages on WSIs to determine consensus labels. Furthermore, 1850 image patches were randomly extracted from Cohort 1 WSIs and binarily classified for tumor viability. A deep-learning model employing knowledge distillation was developed to automatically classify positive patches for each WSI and estimate the viable residual tumor percentages. Spatial heatmaps were output for model explanations and visualizations. The approach achieved high concordance with pathologist consensus, with an R^2 of 0.8437, a RAcc_0.1 of 0.7586, a RAcc_0.3 of 0.9885, which were comparable to two senior pathologists (R^2 of 0.9202/0.9619, RAcc_0.1 of 8506/0.9425, RAcc_0.3 of 1.000/1.000) and surpassing two junior pathologists (R^2 of 0.5592/0.5474, RAcc_0.1 of 0.5287/0.5287, RAcc_0.3 of 0.9080/0.9310). Visualizations enabled the localization of residual viable tumor to augment microscopic assessment. This work illustrates deep learning's potential for assisting pathological response assessment. Spatial heatmaps and patch examples provide intuitive explanations of model predictions, engendering clinical trust and adoption(Code and data will be available at https://github.com/WinnieLaugh/ESCC_Percentage once the paper has been conditionally accepted). Integrating interpretable computational pathology could help enhancethe efficiency and consistency of tumor response assessment and empower precise oncology treatment decisions.

Modified Adaptive CNN for Deep Learning based Histopathological Image Analysis for Cancer Diagnosis

This research on cancer diagnosis is motivated by the need for more accurate and efficient diagnostic tools in the field of oncology. Traditional methods of histopathological examination are time-consuming and can be subject to human error. The adoption of deep learning techniques, particularly Convolutional Neural Networks (CNNs), presents an opportunity to significantly improve the precision and speed of cancer diagnosis through automated analysis of tissue samples. This research addresses current challenges, including variability in samples and limited dataset diversity, with the goal of enhancing the reliability and applicability of the developed models. The ultimate aim is to contribute to the advancement of diagnostic methodologies, providing clinicians with more reliable tools for early and accurate cancer detection and ultimately improving patient outcomes and treatment strategies. Using a modified adaptive CNN algorithm, trained on extensive pathology image datasets, the approach enhances the precision and speed of cancer cell detection in tissue samples. By utilizing these artificial intelligence tools, the diagnostic process becomes not only more accurate but also significantly expeditious. Despite the promising strides, certain challenges persist. Variability in tissue samples, limited availability of diverse datasets, and interpretability of deep learning models pose hurdles to widespread adoption. To address these challenges, this research focuses on developing robust models that can generalize well across diverse datasets. Additionally, efforts are being made to create more comprehensive and diverse datasets to improve model training. Results show accuracy exceeding 97%, sensitivity and specificity at 97% and 95%, F1 score reaching 96%, and precision at 96%.

Modified Adaptive CNN for Deep Learning based Histopathological Image Analysis for Cancer Diagnosis

This research on cancer diagnosis is motivated by the need for more accurate and efficient diagnostic tools in the field of oncology. Traditional methods of histopathological examination are time-consuming and can be subject to human error. The adoption of deep learning techniques, particularly Convolutional Neural Networks (CNNs), presents an opportunity to significantly improve the precision and speed of cancer diagnosis through automated analysis of tissue samples. This research addresses current challenges, including variability in samples and limited dataset diversity, with the goal of enhancing the reliability and applicability of the developed models. The ultimate aim is to contribute to the advancement of diagnostic methodologies, providing clinicians with more reliable tools for early and accurate cancer detection and ultimately improving patient outcomes and treatment strategies. Using a modified adaptive CNN algorithm, trained on extensive pathology image datasets, the approach enhances the precision and speed of cancer cell detection in tissue samples. By utilizing these artificial intelligence tools, the diagnostic process becomes not only more accurate but also significantly expeditious. Despite the promising strides, certain challenges persist. Variability in tissue samples, limited availability of diverse datasets, and interpretability of deep learning models pose hurdles to widespread adoption. To address these challenges, this research focuses on developing robust models that can generalize well across diverse datasets. Additionally, efforts are being made to create more comprehensive and diverse datasets to improve model training. Results show accuracy exceeding 97%, sensitivity and specificity at 97% and 95%, F1 score reaching 96%, and precision at 96%.

How Artificial Intelligence Unravels the Complex Web of Cancer Drug Response.

The intersection of precision medicine and artificial intelligence (AI) holds profound implications for cancer treatment, with the potential to significantly advance our understanding of drug responses based on the intricate architecture of tumor cells. A recent study by Park and colleagues titled "A Deep Learning Model of Tumor Cell Architecture Elucidates Response and Resistance to CDK4/6 Inhibitors" epitomizes this intersection by leveraging an interpretable deep learning model grounded in a comprehensive map of multiprotein assemblies in cancer, known as Nested Systems in Tumors. This study not only elucidates mechanisms underlying the response to CDK4/6 inhibitors in breast cancer therapy but also highlights the critical role of model interpretability leading to new mechanistic insights.

Deep learning of pretreatment multiphase CT images for predicting response to lenvatinib and immune checkpoint inhibitors in unresectable hepatocellular carcinoma

ObjectivesCombination therapy of lenvatinib and immune checkpoint inhibitors (CLICI) has emerged as a promising approach for managing unresectable hepatocellular carcinoma (HCC). However, the response to such treatment is observed in only a subset of patients, underscoring the pressing need for reliable methods to identify potential responders. Materials & methodsThis was a retrospective analysis involving 120 patients with unresectable HCC. They were divided into training (n = 72) and validation (n = 48) cohorts. We developed an interpretable deep learning model using multiphase computed tomography (CT) images to predict whether patients will respond or not to CLICI treatment, based on the Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST v1.1). We evaluated the models' performance and analyzed the impact of each CT phase. Critical regions influencing predictions were identified and visualized through heatmaps. ResultsThe multiphase model outperformed the best biphase and uniphase models, achieving an area under the curve (AUC) of 0.802 (95% CI = 0.780–0.824). The portal phase images were found to significantly enhance the model's predictive accuracy. Heatmaps identified six critical features influencing treatment response, offering valuable insights to clinicians. Additionally, we have made this model accessible via a web server at http://uhccnet.com/ for ease of use. ConclusionsThe integration of multiphase CT images with deep learning-generated heatmaps for predicting treatment response provides a robust and practical tool for guiding CLICI therapy in patients with unresectable HCC.

Smarter water quality monitoring in reservoirs using interpretable deep learning models and feature importance analysis

This study utilized datasets from an ongoing monitoring project conducted in Wadi Dayqah Dam, the largest reservoir in Oman. The dataset comprises information on ten water quality variables collected by AAQ-RINKO device for 20 field sampling stations, encompassing different depths within the water columns. First, an in-depth data analysis was conducted to process and characterize variations in key parameters. Then, four deep learning models were developed and evaluated, including the Gated Residual Variable Selection (GRVS), Deep and Cross (DC), Deep and Wide (DW), and Base models for estimating two important water quality parameters, namely dissolved oxygen (DO) and chlorophyll-a (Chl-a). To justify the performance of deep learning methods, they were compared to four traditional machine learning techniques: MLP, SVR, AdaBoost, and LR-Lasso models. The GRVS model emerged as the most effective method, achieving high accuracy with R2 and RMSE values of 0.98 and 0.38 for the DO parameter and 0.84, and 0.45 for the Chl-a parameter, respectively. Finally, two SHapley Additive exPlanations (SHAP) and Softmax layer model interpretation approaches were employed to highlight the influential factors affecting the predictions in the best model. Results from the SHAP analysis identified pH, depth, and temperature as the most significant variables, with mean SHAP values of 1.8, 0.75, and 0.36 for DO and 0.74, 0.37, and 0.3 for Chl-a, respectively. The data-driven framework implemented in this study holds promise for efficiently approximating hard-to-measure water quality indicators in monitoring projects using cost-effective inputs, which is particularly valuable in resource-constrained settings.

An integrated framework for prognosis prediction and drug response modeling in colorectal liver metastasis drug discovery

BackgroundColorectal cancer (CRC) is the third most prevalent cancer globally, and liver metastasis (CRLM) is the primary cause of death. Hence, it is essential to discover novel prognostic biomarkers and therapeutic drugs for CRLM.MethodsThis study developed two liver metastasis-associated prognostic signatures based on differentially expressed genes (DEGs) in CRLM. Additionally, we employed an interpretable deep learning model utilizing drug sensitivity databases to identify potential therapeutic drugs for high-risk CRLM patients. Subsequently, in vitro and in vivo experiments were performed to verify the efficacy of these compounds.ResultsThese two prognostic models exhibited superior performance compared to previously reported ones. Obatoclax, a BCL-2 inhibitor, showed significant differential responses between high and low risk groups classified by prognostic models, and demonstrated remarkable effectiveness in both Transwell assay and CT26 colorectal liver metastasis mouse model.ConclusionsThis study highlights the significance of developing specialized prognostication approaches and investigating effective therapeutic drugs for patients with CRLM. The application of a deep learning drug response model provides a new drug discovery strategy for translational medicine in precision oncology.

Abstract 6189: Deep-learning model trained on multiplex immunofluorescence-stained tissue samples predicts the survival of patients with non-small cell lung cancer better than PD-L1 TPS alone

Abstract In non-small cell lung cancer (NSCLC), the composition and structure of the tumor immune microenvironment (TME) critically impact patient outcomes, necessitating improved computational approaches to characterize the TME beyond traditional metrics like the PD-L1 tumor proportion score (TPS). Here, we developed an interpretable deep-learning model that robustly predicts patient overall survival (OS) and identifies unique PD-L1 enriched neighborhoods linked to patient prognosis. We utilized a dataset of 507 NSCLC patient biopsy samples prospectively acquired as part of the DFCI ImmunoPROFILE project. Each sample was stained for Cytokeratin as tumor marker, PD-L1, PD-1, CD8, and FOXP3 using a targeted multiplex immunofluorescence assay. Inner tumor regions of interest were automatically processed for spatially resolved identification and quantification of marker expression at the single-cell level. The dataset was divided into 380 patients for model development and 127 for testing. The deep-learning model, a graph neural network (GNN), was trained to predict OS based on local cell neighborhoods using a weakly supervised training paradigm. The GNN models neighborhoods as a graph, with cells and their corresponding marker expression representing nodes and edges connecting adjacent cells, for an average of 42 cells per neighborhood. Survival predictions were obtained by averaging predictions over all neighborhoods from one sample. The GNN model demonstrated robust performance with a concordance index (c-index) of 0.79 on the test data, surpassing traditional metrics such as TPS and immune marker density-based models (c-index: 0.50 and 0.74, respectively). Beyond assessing performance, we investigated the features learned by the GNN using k-means clustering in the model’s feature space. We identified three clusters that were notably enriched in PD-L1 positive tumor cells. The first cluster shows an especially “high TPS” phenotype, where the GNN predicted such neighborhoods to be generally poor for survival. The other two clusters were additionally enriched with PD-L1 positive immune cells (ICs) and associated with favorable survival predictions. Notably, one cluster was also enriched with other types of ICs (”PD-L1 IC mixed”), which the GNN predicted as especially favorable for survival compared to exclusive enrichment of PD-L1 ICs (”PD-L1 IC dominant”). A score summarizing these three PD-L1 expression patterns was significantly associated with OS (p-value: 0.01; HR: 0.72, 95% CI: 0.16-1.27), validating the features learned by the GNN. In sum, we developed a highly interpretable deep-learning model leveraging cellular neighborhood graphs, which effectively identifies expression patterns correlating with NSCLC prognosis, offering promise for developing personalized treatment strategies. Citation Format: Katharina Viktoria Hoebel, James R. Lindsay, Joao V. Alessi, Jason L. Weirather, Ian D. Dryg, Jennifer Altreuter, Mark M. Awad, Scott J. Rodig, William E. Lotter. Deep-learning model trained on multiplex immunofluorescence-stained tissue samples predicts the survival of patients with non-small cell lung cancer better than PD-L1 TPS alone [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6189.

Explaining deep learning models for ozone pollution prediction via embedded feature selection

Ambient air pollution is a pervasive global issue that poses significant health risks. Among pollutants, ozone (O3) is responsible for an estimated 1 to 1.2 million premature deaths yearly. Furthermore, O3 adversely affects climate warming, crop productivity, and more. Its formation occurs when nitrogen oxides and volatile organic compounds react with short-wavelength solar radiation. Consequently, urban areas with high traffic volume and elevated temperatures are particularly prone to elevated O3 levels, which pose a significant health risk to their inhabitants. In response to this problem, many countries have developed web and mobile applications that provide real-time air pollution information using sensor data. However, while these applications offer valuable insight into current pollution levels, predicting future pollutant behavior is crucial for effective planning and mitigation strategies. Therefore, our main objectives are to develop accurate and efficient prediction models and identify the key factors that influence O3 levels. We adopt a time series forecasting approach to address these objectives, which allows us to analyze and predict future O3 behavior. Additionally, we tackle the feature selection problem to identify the most relevant features and periods that contribute to prediction accuracy by introducing a novel method called the Time Selection Layer in Deep Learning models, which significantly improves model performance, reduces complexity, and enhances interpretability. Our study focuses on data collected from five representative areas in Seville, Cordova, and Jaen provinces in Spain, using multiple sensors to capture comprehensive pollution data. We compare the performance of three models: Lasso, Decision Tree, and Deep Learning with and without incorporating the Time Selection Layer. Our results demonstrate that including the Time Selection Layer significantly enhances the effectiveness and interpretability of Deep Learning models, achieving an average effectiveness improvement of 9% across all monitored areas.

Exploring the Application of Machine Learning in the Field of Fruit Image Recognition

Image recognition techniques play an important role in business and science fields. The application of these techniques, especially the YOLO algorithm's real-time processing capabilities and high accuracy make it particularly noteworthy in advancing fruit image recognition. This study focuses on the application of machine learning in the field of image recognition, and in particular explores the use of the YOLO (You Only Look Once) algorithm in fruit image recognition. The study experimentally explores the application of the YOLO algorithm to a complex fruit recognition task, with a focus on evaluating the algorithm's real-time processing power and accuracy. Using a wide range of fruit image datasets including COCO, OID and KITTI, the YOLO model was trained and tested with the aim of gaining insights into the algorithm's performance in different real-world scenarios and its limitations. This research points out the importance of continuous model improvement, especially in terms of image quality, model adaptation, and with other techniques such as NLP. Directions for future research include the development of more transparent and interpretable deep learning models.

Predicting long-term progression of Alzheimer’s disease using a multimodal deep learning model incorporating interaction effects

BackgroundIdentifying individuals with mild cognitive impairment (MCI) at risk of progressing to Alzheimer’s disease (AD) provides a unique opportunity for early interventions. Therefore, accurate and long-term prediction of the conversion from MCI to AD is desired but, to date, remains challenging. Here, we developed an interpretable deep learning model featuring a novel design that incorporates interaction effects and multimodality to improve the prediction accuracy and horizon for MCI-to-AD progression.MethodsThis multi-center, multi-cohort retrospective study collected structural magnetic resonance imaging (sMRI), clinical assessments, and genetic polymorphism data of 252 patients with MCI at baseline from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Our deep learning model was cross-validated on the ADNI-1 and ADNI-2/GO cohorts and further generalized in the ongoing ADNI-3 cohort. We evaluated the model performance using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and F1 score.ResultsOn the cross-validation set, our model achieved superior results for predicting MCI conversion within 4 years (AUC, 0.962; accuracy, 92.92%; sensitivity, 88.89%; specificity, 95.33%) compared to all existing studies. In the independent test, our model exhibited consistent performance with an AUC of 0.939 and an accuracy of 92.86%. Integrating interaction effects and multimodal data into the model significantly increased prediction accuracy by 4.76% (P = 0.01) and 4.29% (P = 0.03), respectively. Furthermore, our model demonstrated robustness to inter-center and inter-scanner variability, while generating interpretable predictions by quantifying the contribution of multimodal biomarkers.ConclusionsThe proposed deep learning model presents a novel perspective by combining interaction effects and multimodality, leading to more accurate and longer-term predictions of AD progression, which promises to improve pre-dementia patient care.