Explore the vast range of titles available.

Deep neural networks have been well-known for their superb handling of various machine learning and artificial intelligence tasks. However, due to their over-parameterized black-box nature, it is often difficult to understand the prediction results of deep models. In recent years, many interpretation tools have been proposed to explain or reveal how deep models make decisions. In this paper, we review this line of research and try to make a comprehensive survey. Specifically, we first introduce and clarify two basic concepts—interpretations and interpretability—that people usually get confused about. To address the research efforts in interpretations, we elaborate the designs of a number of interpretation algorithms, from different perspectives, by proposing a new taxonomy. Then, to understand the interpretation results, we also survey the performance metrics for evaluating interpretation algorithms. Further, we summarize the current works in evaluating models’ interpretability using “trustworthy” interpretation algorithms. Finally, we review and discuss the connections between deep models’ interpretations and other factors, such as adversarial robustness and learning from interpretations, and we introduce several open-source libraries for interpretation algorithms and evaluation approaches.

Alzheimer’s Disease (AD) pathology has been increasingly explored through single-cell and single-nucleus RNA-sequencing (scRNA-seq & snRNA-seq) and spatial transcriptomics (ST). However, the surge in data demands a comprehensive, user-friendly repository. Addressing this, we introduce a single-cell and spatial RNA-seq database for Alzheimer’s disease (ssREAD). It offers a broader spectrum of AD-related datasets, an optimized analytical pipeline, and improved usability. The database encompasses 1,053 samples (277 integrated datasets) from 67 AD-related scRNA-seq & snRNA-seq studies, totaling 7,332,202 cells. Additionally, it archives 381 ST datasets from 18 human and mouse brain studies. Each dataset is annotated with details such as species, gender, brain region, disease/control status, age, and AD Braak stages. ssREAD also provides an analysis suite for cell clustering, identification of differentially expressed and spatially variable genes, cell-type-specific marker genes and regulons, and spot deconvolution for integrative analysis. ssREAD is freely available at https://bmblx.bmi.osumc.edu/ssread/ . A systematic collection for single-cell and spatial transcriptomics is critical for in-depth analysis and novel discovery in AD. Here, authors show ssREAD which covers over 7 million cells and 381 spatial samples from human and mouse, providing a comprehensive resource for AD research.

We propose a variant of residual networks (ResNets) for galaxy morphology classification. The variant, together with other popular convolutional neural networks (CNNs), is applied to a sample of 28790 galaxy images from the Galaxy Zoo 2 dataset, to classify galaxies into five classes, i.e., completely round smooth, in-between smooth (between completely round and cigar-shaped), cigar-shaped smooth, edge-on and spiral. Various metrics, such as accuracy, precision, recall, F1 value and AUC, show that the proposed network achieves state-of-the-art classification performance among other networks, namely, Dieleman, AlexNet, VGG, Inception and ResNets. The overall classification accuracy of our network on the testing set is 95.2083% and the accuracy of each type is given as follows: completely round, 96.6785%; in-between, 94.4238%; cigar-shaped, 58.6207%; edge-on, 94.3590% and spiral, 97.6953%. Our model algorithm can be applied to large-scale galaxy classification in forthcoming surveys, such as the Large Synoptic Survey Telescope (LSST) survey.

Background Nowadays, trendy research in biomedical sciences juxtaposes the term ‘precision’ to medicine and public health with companion words like big data, data science, and deep learning. Technological advancements permit the collection and merging of large heterogeneous datasets from different sources, from genome sequences to social media posts or from electronic health records to wearables. Additionally, complex algorithms supported by high-performance computing allow one to transform these large datasets into knowledge. Despite such progress, many barriers still exist against achieving precision medicine and precision public health interventions for the benefit of the individual and the population. Main body The present work focuses on analyzing both the technical and societal hurdles related to the development of prediction models of health risks, diagnoses and outcomes from integrated biomedical databases. Methodological challenges that need to be addressed include improving semantics of study designs: medical record data are inherently biased, and even the most advanced deep learning’s denoising autoencoders cannot overcome the bias if not handled a priori by design. Societal challenges to face include evaluation of ethically actionable risk factors at the individual and population level; for instance, usage of gender, race, or ethnicity as risk modifiers, not as biological variables, could be replaced by modifiable environmental proxies such as lifestyle and dietary habits, household income, or access to educational resources. Conclusions Data science for precision medicine and public health warrants an informatics-oriented formalization of the study design and interoperability throughout all levels of the knowledge inference process, from the research semantics, to model development, and ultimately to implementation.

Heat barrier, the unrestricted increase in airplane or rocket speeds caused by aerodynamic heating, which—without adequate provisions for cooling the exposed surfaces—can lead to the loss of a hypersonic vehicle’s reusability, maneuverability, and cost-effectiveness. To date, indirect thermal protection methods, such as regenerative cooling, film cooling, and transpiration cooling, have proven to be complex and inefficient. Here, we propose a direct liquid cooling system to mitigate the heat barrier, utilizing a blunt-sharp structured thermal armor (STA)—a recently proposed material [36] to elevate the Leidenfrost point. The fiber-metal nano-/micro-STA withstands rigorous simulated hypersonic aerodynamic heating using butane and acetylene flames, ensuring effective temperature management in scenarios where flame temperatures reach up to 3000 °C—far exceeding the melting point of the STA substrate. Systematic cycling and durability tests further confirm the STA’s exceptional tolerance and robustness under extreme conditions. This work offers an efficient thermal protection method for hypersonic vehicles. Heat barriers pose significant challenges to hypersonic flight. Here, authors demonstrate a direct liquid cooling system using a structured thermal armor that elevates the Leidenfrost point, effectively managing temperatures up to 3000 °C.

There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. However, there are few clinical language models, the largest of which trained in the clinical domain is comparatively small at 110 million parameters (compared with billions of parameters in the general domain). It is not clear how large clinical language models with billions of parameters can help medical AI systems utilize unstructured EHRs. In this study, we develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text) and systematically evaluate it on five clinical NLP tasks including clinical concept extraction, medical relation extraction, semantic textual similarity, natural language inference (NLI), and medical question answering (MQA). We examine how (1) scaling up the number of parameters and (2) scaling up the size of the training data could benefit these NLP tasks. GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks (e.g., 9.6% and 9.5% improvement in accuracy for NLI and MQA), which can be applied to medical AI systems to improve healthcare delivery. The GatorTron models are publicly available at: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/models/gatortron_og .

There are enormous enthusiasm and concerns in applying large language models (LLMs) to healthcare. Yet current assumptions are based on general-purpose LLMs such as ChatGPT, which are not developed for medical use. This study develops a generative clinical LLM, GatorTronGPT, using 277 billion words of text including (1) 82 billion words of clinical text from 126 clinical departments and approximately 2 million patients at the University of Florida Health and (2) 195 billion words of diverse general English text. We train GatorTronGPT using a GPT-3 architecture with up to 20 billion parameters and evaluate its utility for biomedical natural language processing (NLP) and healthcare text generation. GatorTronGPT improves biomedical natural language processing. We apply GatorTronGPT to generate 20 billion words of synthetic text. Synthetic NLP models trained using synthetic text generated by GatorTronGPT outperform models trained using real-world clinical text. Physicians’ Turing test using 1 (worst) to 9 (best) scale shows that there are no significant differences in linguistic readability (p = 0.22; 6.57 of GatorTronGPT compared with 6.93 of human) and clinical relevance (p = 0.91; 7.0 of GatorTronGPT compared with 6.97 of human) and that physicians cannot differentiate them (p < 0.001). This study provides insights into the opportunities and challenges of LLMs for medical research and healthcare.

Big data, high-performance computing, and (deep) machine learning are increasingly becoming key to precision medicine—from identifying disease risks and taking preventive measures, to making diagnoses and personalizing treatment for individuals. Precision medicine, however, is not only about predicting risks and outcomes, but also about weighing interventions. Interventional clinical predictive models require the correct specification of cause and effect, and the calculation of so-called counterfactuals, that is, alternative scenarios. In biomedical research, observational studies are commonly affected by confounding and selection bias. Without robust assumptions, often requiring a priori domain knowledge, causal inference is not feasible. Data-driven prediction models are often mistakenly used to draw causal effects, but neither their parameters nor their predictions necessarily have a causal interpretation. Therefore, the premise that data-driven prediction models lead to trustable decisions/interventions for precision medicine is questionable. When pursuing intervention modelling, the bio-health informatics community needs to employ causal approaches and learn causal structures. Here we discuss how target trials (algorithmic emulation of randomized studies), transportability (the licence to transfer causal effects from one population to another) and prediction invariance (where a true causal model is contained in the set of all prediction models whose accuracy does not vary across different settings) are linchpins to developing and testing intervention models. Machine learning models are commonly used to predict risks and outcomes in biomedical research. But healthcare often requires information about cause–effect relations and alternative scenarios, that is, counterfactuals. Prosperi et al. discuss the importance of interventional and counterfactual models, as opposed to purely predictive models, in the context of precision medicine.

Irrigation water management and real-time monitoring of crop water stress status can enhance agricultural water use efficiency, crop yield, and crop quality. The aim of this study was to simplify the calculation of the crop water stress index (CWSI) and improve its diagnostic accuracy. Simplified CWSI (CWSIsi) was used to diagnose water stress for cotton that has received four different irrigation treatments (no stress, mild stress, moderate stress, and severe stress) at the flowering and boll stage. High resolution thermal infrared and multispectral images were taken using an Unmanned Aerial Vehicle remote sensing platform at midday (local time 13:00), and stomatal conductance (gs), transpiration rate (tr), and cotton root zone soil volumetric water content (θ) were concurrently measured. The soil background pixels of thermal images were eliminated using the Canny edge detection to obtain a unimodal histogram of pure canopy temperatures. Then the wet reference temperature (Twet), dry reference temperature (Tdry), and mean canopy temperature (Tl) were obtained from the canopy temperature histogram to calculate CWSIsi. The other two methods of CWSI evaluation were empirical CWSI (CWSIe), in which the temperature parameters were determined by measuring natural reference cotton leaves, and statistical CWSI (CWSIs), in which Twet was the mean of the lowest 5% of canopy temperatures and Tdry was the air temperature (Tair) + 5 °C. Compared with CWSIe, CWSIs and spectral indices (NDVI, TCARI, OSAVI, TCARI/OSAVI), CWSIsi has higher correlation with gs (R2 = 0.660) and tr (R2 = 0.592). The correlation coefficient (R) for θ (0–45 cm) and CWSIsi is also high (0.812). The plotted high-resolution map of CWSIsi shows the different distribution of cotton water stress in different irrigation treatments. These findings demonstrate that CWSIsi, which only requires parameters from a canopy temperature histogram, may potentially be applied to precision irrigation management.

Background The Short Form-8 (SF-8) is a widely used instrument for measuring health-related quality of life (HRQOL). The purpose of the current study is to evaluate the reliability and validity of the Chinese version SF-8 using a large, representative sample of city residents in mainland China. Methods We surveyed residents of 35 major cities in China using random digit dialing of both landlines and cell phones. We adopted a multi-stage stratified sampling scheme and selected a probability sample of 10,885 adults. Internal consistency reliability of the SF-8 was evaluated with item-total correlations and Cronbach’s alphas. Construct validity was assessed with factor analysis. Known-groups validity was examined based on known HRQOL differences in age, gender, income, and overall quality of life. Results We showed that SF-8 has very good internal consistency reliability and known-groups validity. Our results also confirmed that the traditional 2-factor structure of SF-8 (physical and mental health) is reasonable among Chinese city residents. Further, we showed that a 3-factor model (physical, mental, and overall health) fit the data better than the traditional 2-factor model. Conclusions This study is the first to confirm the traditional 2-factor structure of SF-8 using a large, representative sample from China. We have shown that the SF-8 Chinese version is feasible, reliable, and valid. Our findings support the use of the SF-8 summary scores for assessing general HRQOL among Chinese. Future studies may further explore the possibility of a 3-factor structure for the SF-8 among the Chinese population.