Infrared Dim Small Target Detection Networks: A Review.

In recent years, with the rapid development of deep learning and its outstanding capabilities in target detection, innovative methods have been introduced for infrared dim small target detection. This review comprehensively summarizes public datasets, the latest networks, and evaluation metrics for infrared dim small target detection. This review mainly focuses on deep learning methods from the past three years and categorizes them based on the six key issues in this field: (1) enhancing the representation capability of small targets; (2) improving the accuracy of bounding box regression; (3) resolving the issue of target information loss in the deep network; (4) balancing missed detections and false alarms; (5) adapting for complex backgrounds; (6) lightweight design and deployment issues of the network. Additionally, this review summarizes twelve public datasets for infrared dim small targets and evaluation metrics used for detection and quantitatively compares the performance of the latest networks. Finally, this review provides insights into the future directions of this field. In conclusion, this review aims to assist researchers in gaining a comprehensive understanding of the latest developments in infrared dim small target detection networks.

Deep learning models for ischemic stroke lesion segmentation in medical images: A survey.

This paper provides a comprehensive review of deep learning models for ischemic stroke lesion segmentation in medical images. Ischemic stroke is a severe neurological disease and a leading cause of death and disability worldwide. Accurate segmentation of stroke lesions in medical images such as MRI and CT scans is crucial for diagnosis, treatment planning and prognosis. This paper first introduces common imaging modalities used for stroke diagnosis, discussing their capabilities in imaging lesions at different disease stages from the acute to chronic stage. It then reviews three major public benchmark datasets for evaluating stroke segmentation algorithms: ATLAS, ISLES and AISD, highlighting their key characteristics. The paper proceeds to provide an overview of foundational deep learning architectures for medical image segmentation, including CNN-based and transformer-based models. It summarizes recent innovations in adapting these architectures to the task of stroke lesion segmentation across the three datasets, analyzing their motivations, modifications and results. A survey of loss functions and data augmentations employed for this task is also included. The paper discusses various aspects related to stroke segmentation tasks, including prior knowledge, small lesions, and multimodal fusion, and then concludes by outlining promising future research directions. Overall, this comprehensive review covers critical technical developments in the field to support continued progress in automated stroke lesion segmentation.

Public data homogenization for AI model development in breast cancer.

BACKGROUND: Developing trustworthy artificial intelligence (AI) models for clinical applications requires access to clinical and imaging data cohorts. Reusing of publicly available datasets has the potential to fill this gap. Specifically in the domain of breast cancer, a large archive of publicly accessible medical images along with the corresponding clinical data is available at The Cancer Imaging Archive (TCIA). However, existing datasets cannot be directly used as they are heterogeneous and cannot be effectively filtered for selecting specific image types required to develop AI models. This work focuses on the development of a homogenized dataset in the domain of breast cancer including clinical and imaging data. METHODS: Five datasets were acquired from the TCIA and were harmonized. For the clinical data harmonization, a common data model was developed and a repeatable, documented "extract-transform-load" process was defined and executed for their homogenization. Further, Digital Imaging and COmmunications in Medicine (DICOM) information was extracted from magnetic resonance imaging (MRI) data and made accessible and searchable. RESULTS: The resulting harmonized dataset includes information about 2,035 subjects with breast cancer. Further, a platform named RV-Cherry-Picker enables search over both the clinical and diagnostic imaging datasets, providing unified access, facilitating the downloading of all study imaging that correspond to specific series' characteristics (e.g., dynamic contrast-enhanced series), and reducing the burden of acquiring the appropriate set of images for the respective AI model scenario. CONCLUSIONS: RV-Cherry-Picker provides access to the largest, publicly available, homogenized, imaging/clinical dataset for breast cancer to develop AI models on top. RELEVANCE STATEMENT: We present a solution for creating merged public datasets supporting AI model development, using as an example the breast cancer domain and magnetic resonance imaging images. KEY POINTS: • The proposed platform allows unified access to the largest, homogenized public imaging dataset for breast cancer. • A methodology for the semantically enriched homogenization of public clinical data is presented. • The platform is able to make a detailed selection of breast MRI data for the development of AI models.

Combining public datasets for automated tooth assessment in panoramic radiographs.

OBJECTIVE: Panoramic radiographs (PRs) provide a comprehensive view of the oral and maxillofacial region and are used routinely to assess dental and osseous pathologies. Artificial intelligence (AI) can be used to improve the diagnostic accuracy of PRs compared to bitewings and periapical radiographs. This study aimed to evaluate the advantages and challenges of using publicly available datasets in dental AI research, focusing on solving the novel task of predicting tooth segmentations, FDI numbers, and tooth diagnoses, simultaneously. MATERIALS AND METHODS: Datasets from the OdontoAI platform (tooth instance segmentations) and the DENTEX challenge (tooth bounding boxes with associated diagnoses) were combined to develop a two-stage AI model. The first stage implemented tooth instance segmentation with FDI numbering and extracted regions of interest around each tooth segmentation, whereafter the second stage implemented multi-label classification to detect dental caries, impacted teeth, and periapical lesions in PRs. The performance of the automated tooth segmentation algorithm was evaluated using a free-response receiver-operating-characteristics (FROC) curve and mean average precision (mAP) metrics. The diagnostic accuracy of detection and classification of dental pathology was evaluated with ROC curves and F1 and AUC metrics. RESULTS: The two-stage AI model achieved high accuracy in tooth segmentations with a FROC score of 0.988 and a mAP of 0.848. High accuracy was also achieved in the diagnostic classification of impacted teeth (F1 = 0.901, AUC = 0.996), whereas moderate accuracy was achieved in the diagnostic classification of deep caries (F1 = 0.683, AUC = 0.960), early caries (F1 = 0.662, AUC = 0.881), and periapical lesions (F1 = 0.603, AUC = 0.974). The model's performance correlated positively with the quality of annotations in the used public datasets. Selected samples from the DENTEX dataset revealed cases of missing (false-negative) and incorrect (false-positive) diagnoses, which negatively influenced the performance of the AI model. CONCLUSIONS: The use and pooling of public datasets in dental AI research can significantly accelerate the development of new AI models and enable fast exploration of novel tasks. However, standardized quality assurance is essential before using the datasets to ensure reliable outcomes and limit potential biases.

The University of California San Francisco Brain Metastases Stereotactic Radiosurgery (UCSF-BMSR) MRI Dataset.

Supplemental material is available for this article.

Deep learning for computer-aided abnormalities classification in digital mammogram: A data-centric perspective.

Breast cancer is the most common type of cancer in women, and early abnormality detection using mammography can significantly improve breast cancer survival rates. Diverse datasets are required to improve the training and validation of deep learning (DL) systems for autonomous breast cancer diagnosis. However, only a small number of mammography datasets are publicly available. This constraint has created challenges when comparing different DL models using the same dataset. The primary contribution of this study is the comprehensive description of a selection of currently available public mammography datasets. The information available on publicly accessible datasets is summarized and their usability reviewed to enable more effective models to be developed for breast cancer detection and to improve understanding of existing models trained using these datasets. This study aims to bridge the existing knowledge gap by offering researchers and practitioners a valuable resource to develop and assess DL models in breast cancer diagnosis.

BUS-Set: A benchmark for quantitative evaluation of breast ultrasound segmentation networks with public datasets.

PURPOSE: BUS-Set is a reproducible benchmark for breast ultrasound (BUS) lesion segmentation, comprising of publicly available images with the aim of improving future comparisons between machine learning models within the field of BUS. METHOD: Four publicly available datasets were compiled creating an overall set of 1154 BUS images, from five different scanner types. Full dataset details have been provided, which include clinical labels and detailed annotations. Furthermore, nine state-of-the-art deep learning architectures were selected to form the initial benchmark segmentation result, tested using five-fold cross-validation and MANOVA/ANOVA with Tukey statistical significance test with a threshold of 0.01. Additional evaluation of these architectures was conducted, exploring possible training bias, and lesion size and type effects. RESULTS: Of the nine state-of-the-art benchmarked architectures, Mask R-CNN obtained the highest overall results, with the following mean metric scores: Dice score of 0.851, intersection over union of 0.786 and pixel accuracy of 0.975. MANOVA/ANOVA and Tukey test results showed Mask R-CNN to be statistically significant better compared to all other benchmarked models with a p-value >0.01. Moreover, Mask R-CNN achieved the highest mean Dice score of 0.839 on an additional 16 image dataset, that contained multiple lesions per image. Further analysis on regions of interest was conducted, assessing Hamming distance, depth-to-width ratio (DWR), circularity, and elongation, which showed that the Mask R-CNN's segmentations maintained the most morphological features with correlation coefficients of 0.888, 0.532, 0.876 for DWR, circularity, and elongation, respectively. Based on the correlation coefficients, statistical test indicated that Mask R-CNN was only significantly different to Sk-U-Net. CONCLUSIONS: BUS-Set is a fully reproducible benchmark for BUS lesion segmentation obtained through the use of public datasets and GitHub. Of the state-of-the-art convolution neural network (CNN)-based architectures, Mask R-CNN achieved the highest performance overall, further analysis indicated that a training bias may have occurred due to the lesion size variation in the dataset. All dataset and architecture details are available at GitHub: https://github.com/corcor27/BUS-Set, which allows for a fully reproducible benchmark.

Computer vision-based solutions to overcome the limitations of wireless capsule endoscopy.

Endoscopic investigation plays a critical role in the diagnosis of gastrointestinal (GI) diseases. Since 2001, Wireless Capsule Endoscopy (WCE) has been available for small bowel exploration and is in continuous development. Over the last decade, WCE has achieved impressive improvements in areas such as miniaturisation, image quality and battery life. As a result, WCE is currently a very useful alternative to wired enteroscopy in the investigation of various small bowel abnormalities and has the potential to become the leading screening technique for the entire gastrointestinal tract. However, commercial solutions still have several limitations, namely incomplete examination and limited diagnostic capacity. These deficiencies are related to technical issues, such as image quality, motion estimation and power consumption management. Computational methods, based on image processing and analysis, can help to overcome these challenges and reduce both the time required by reviewers and human interpretation errors. Research groups have proposed a series of methods including algorithms for locating the capsule or lesion, assessing intestinal motility and improving image quality.In this work, we provide a critical review of computational vision-based methods for WCE image analysis aimed at overcoming the technological challenges of capsules. This article also reviews several representative public datasets used to evaluate the performance of WCE techniques and methods. Finally, some promising solutions of computational methods based on the analysis of multiple-camera endoscopic images are presented.

Sensitivity of modeled residential fine particulate matter exposure to select building and source characteristics: A case study using public data in Boston, MA.

Many techniques for estimating exposure to airborne contaminants do not account for building characteristics that can magnify contaminant contributions from indoor and outdoor sources. Building characteristics that influence exposure can be challenging to obtain at scale, but some may be incorporated into exposure assessments using public datasets. We present a methodology for using public datasets to generate housing models for a test cohort, and examined sensitivity of predicted fine particulate matter (PM2.5) exposures to selected building and source characteristics. We used addresses of a cohort of children with asthma and public tax assessor's data to guide selection of floorplans of US residences from a public database. This in turn guided generation of coupled multi-zone models (CONTAM and EnergyPlus) that estimated indoor PM2.5 exposure profiles. To examine sensitivity to model parameters, we varied building floors and floorplan, heating, ventilating and air-conditioning (HVAC) type, room or floor-level model resolution, and indoor source strength and schedule (for hypothesized gas stove cooking and tobacco smoking). Occupant time-activity and ambient pollutant levels were held constant. Our address matching methodology identified two multi-family house templates and one single-family house template that had similar characteristics to 60 % of test addresses. Exposure to infiltrated ambient PM2.5 was similar across selected building characteristics, HVAC types, and model resolutions (holding all else equal). By comparison, exposures to indoor-sourced PM2.5 were higher in the two multi-family residences than the single family residence (e.g., for cooking PM2.5 exposure, by 26 % and 47 % respectively) and were sensitive to HVAC type and model resolution. We derived the influence of building characteristics and HVAC type on PM2.5 exposure indoors using public data sources and coupled multi-zone models. With the important inclusion of individualized resident behavior data, similar housing modeling can be used to incorporate exposure variability in health studies of the indoor residential environment.

cdev: a ground-truth based measure to evaluate RNA-seq normalization performance.

Normalization of RNA-seq data has been an active area of research since the problem was first recognized a decade ago. Despite the active development of new normalizers, their performance measures have been given little attention. To evaluate normalizers, researchers have been relying on ad hoc measures, most of which are either qualitative, potentially biased, or easily confounded by parametric choices of downstream analysis. We propose a metric called condition-number based deviation, or cdev, to quantify normalization success. cdev measures how much an expression matrix differs from another. If a ground truth normalization is given, cdev can then be used to evaluate the performance of normalizers. To establish experimental ground truth, we compiled an extensive set of public RNA-seq assays with external spike-ins. This data collection, together with cdev, provides a valuable toolset for benchmarking new and existing normalization methods.

mtDNAcombine: tools to combine sequences from multiple studies.

BACKGROUND: Today an unprecedented amount of genetic sequence data is stored in publicly available repositories. For decades now, mitochondrial DNA (mtDNA) has been the workhorse of genetic studies, and as a result, there is a large volume of mtDNA data available in these repositories for a wide range of species. Indeed, whilst whole genome sequencing is an exciting prospect for the future, for most non-model organisms' classical markers such as mtDNA remain widely used. By compiling existing data from multiple original studies, it is possible to build powerful new datasets capable of exploring many questions in ecology, evolution and conservation biology. One key question that these data can help inform is what happened in a species' demographic past. However, compiling data in this manner is not trivial, there are many complexities associated with data extraction, data quality and data handling. RESULTS: Here we present the mtDNAcombine package, a collection of tools developed to manage some of the major decisions associated with handling multi-study sequence data with a particular focus on preparing sequence data for Bayesian skyline plot demographic reconstructions. CONCLUSIONS: There is now more genetic information available than ever before and large meta-data sets offer great opportunities to explore new and exciting avenues of research. However, compiling multi-study datasets still remains a technically challenging prospect. The mtDNAcombine package provides a pipeline to streamline the process of downloading, curating, and analysing sequence data, guiding the process of compiling data sets from the online database GenBank.

A Large-Scale COVID-19 Twitter Chatter Dataset for Open Scientific Research-An International Collaboration.

As the COVID-19 pandemic continues to spread worldwide, an unprecedented amount of open data is being generated for medical, genetics, and epidemiological research. The unparalleled rate at which many research groups around the world are releasing data and publications on the ongoing pandemic is allowing other scientists to learn from local experiences and data generated on the front lines of the COVID-19 pandemic. However, there is a need to integrate additional data sources that map and measure the role of social dynamics of such a unique worldwide event in biomedical, biological, and epidemiological analyses. For this purpose, we present a large-scale curated dataset of over 1.12 billion tweets, growing daily, related to COVID-19 chatter generated from 1 January 2020 to 27 June 2021 at the time of writing. This data source provides a freely available additional data source for researchers worldwide to conduct a wide and diverse number of research projects, such as epidemiological analyses, emotional and mental responses to social distancing measures, the identification of sources of misinformation, stratified measurement of sentiment towards the pandemic in near real time, among many others.

Use of public datasets in the examination of multimorbidity: Opportunities and challenges.

The interrogation of established, large-scale datasets presents great opportunities in health data science for the linkage and mining of potentially disparate resources to create new knowledge in a fast and cost-efficient manner. The number of datasets that can be queried in the field of multimorbidity is vast, ranging from national administrative and audit datasets, large clinical, technical and biological cohorts, through to more bespoke data collections made available by individual organisations and laboratories. However, with these opportunities also come technical and regulatory challenges that require an informed approach. In this review, we outline the potential benefits of using previously collected data as a vehicle for research activity. We illustrate the added value of combining potentially disparate datasets to find answers to novel questions in the field. We focus on the legal, governance and logistical considerations required to hold and analyse data acquired from disparate sources and outline some of the solutions to these challenges. We discuss the infrastructure resources required and the essential considerations in data curation and informatics management, and briefly discuss some of the analysis approaches currently used.

Crowdsourcing pneumothorax annotations using machine learning annotations on the NIH chest X-ray dataset.

Pneumothorax is a potentially life-threatening condition that requires prompt recognition and often urgent intervention. In the ICU setting, large numbers of chest radiographs are performed and must be interpreted on a daily basis which may delay diagnosis of this entity. Development of artificial intelligence (AI) techniques to detect pneumothorax could help expedite detection as well as localize and potentially quantify pneumothorax. Open image analysis competitions are useful in advancing state-of-the art AI algorithms but generally require large expert annotated datasets. We have annotated and adjudicated a large dataset of chest radiographs to be made public with the goal of sparking innovation in this space. Because of the cumbersome and time-consuming nature of image labeling, we explored the value of using AI models to generate annotations for review. Utilization of this machine learning annotation (MLA) technique appeared to expedite our annotation process with relatively high sensitivity at the expense of specificity. Further research is required to confirm and better characterize the value of MLAs. Our adjudicated dataset is now available for public consumption in the form of a challenge.

Robust Reference Powered Association Test of Genome-Wide Association Studies.

Genome-wide association studies (GWASs) have identified abundant genetic susceptibility loci, GWAS of small sample size are far less from meeting the previous expectations due to low statistical power and false positive results. Effective statistical methods are required to further improve the analyses of massive GWAS data. Here we presented a new statistic (Robust Reference Powered Association Test) to use large public database (gnomad) as reference to reduce concern of potential population stratification. To evaluate the performance of this statistic for various situations, we simulated multiple sets of sample size and frequencies to compute statistical power. Furthermore, we applied our method to several real datasets (psoriasis genome-wide association datasets and schizophrenia genome-wide association dataset) to evaluate the performance. Careful analyses indicated that our newly developed statistic outperformed several previously developed GWAS applications. Importantly, this statistic is more robust than naive merging method in the presence of small control-reference differentiation, therefore likely to detect more association signals.

Computational Analyses Connect Small-Molecule Sensitivity to Cellular Features Using Large Panels of Cancer Cell Lines.

We recently pioneered several analyses of small-molecule sensitivity data collected from large-scale perturbation of hundreds of cancer cell lines with hundreds of small molecules, with cell viability measured as a readout of compound sensitivity. We performed these studies using cancer cell lines previously annotated with cellular, genomic, and basal gene-expression features. By combining small-molecule sensitivity data with these other datasets, we identified new candidate biomarkers of sensitivity, gained insights into small-molecule mechanisms of action, and proposed candidate hypotheses for cancer dependencies (including candidate combination therapies). Nevertheless, given the size of these datasets, we expect that many connections between cellular features and small-molecule sensitivity remain underexplored. In this chapter, we provide a step-by-step account of foundational data-analysis methods underlying our published studies, including working MATLAB code applied to our own public datasets. These procedures will allow others to repeat analyses of our data with new parameters, in additional contexts, and to adapt our procedures to their own datasets.

A comparison of accuracy and computational feasibility of two record linkage algorithms in retrieving vital status information from HIV/AIDS patients registered in Brazilian public databases.

BACKGROUND AND OBJECTIVE: While cross-referencing information from people living with HIV/AIDS (PLWHA) to the official mortality database is a critical step in monitoring the HIV/AIDS epidemic in Brazil, the accuracy of the linkage routine may compromise the validity of the final database, yielding to biased epidemiological estimates. We compared the accuracy and the total runtime of two linkage algorithms applied to retrieve vital status information from PLWHA in Brazilian public databases. METHODS: Nominally identified records from PLWHA were obtained from three distinct government databases. Linkage routines included an algorithm in Python language (PLA) and Reclink software (RlS), a probabilistic software largely utilized in Brazil. Records from PLWHA1 known to be alive were added to those from patients reported as deceased. Data were then searched into the mortality system. Scenarios where 5% and 50% of patients actually dead were simulated, considering both complete cases and 20% missing maternal names. RESULTS: When complete information was available both algorithms had comparable accuracies. In the scenario of 20% missing maternal names, PLA2 and RlS3 had sensitivities of 94.5% and 94.6% (p > 0.5), respectively; after manual reviewing, PLA sensitivity increased to 98.4% (96.6-100.0) exceeding that for RlS (p < 0.01). PLA had higher positive predictive value in 5% death proportion. Manual reviewing was intrinsically required by RlS in up to 14% register for people actually dead, whereas the corresponding proportion ranged from 1.5% to 2% for PLA. The lack of manual inspection did not alter PLA sensitivity when complete information was available. When incomplete data was available PLA sensitivity increased from 94.5% to 98.4%, thus exceeding that presented by RlS (94.6%, p < 0.05). RlS spanned considerably less processing time compared to PLA. CONCLUSION: Both linkage algorithms presented interchangeable accuracies in retrieving vital status data from PLWHA. RlS had a considerably lesser runtime but intrinsically required manually reviewing a fastidious proportion of the matched registries. On the other hand, PLA spent quite more runtime but spared manual reviewing at no expense of accuracy.

Reconstructing the Chernobyl Nuclear Power Plant (CNPP) accident 30 years after. A unique database of air concentration and deposition measurements over Europe.

30 years after the Chernobyl Nuclear Power Plant (CNPP) accident, its radioactive releases still remain of great interest mainly due to the long half-lives of many radionuclides emitted. Observations from the terrestrial environment, which hosts radionuclides for many years after initial deposition, are important for health and environmental assessments. Furthermore, such measurements are the basis for validation of atmospheric transport models and can be used for constraining the still not accurately known source terms. However, although the "Atlas of cesium deposition on Europe after the Chernobyl accident" (hereafter referred to as "Atlas") has been published since 1998, less than 1% of the direct observations of (137)Cs deposition has been made publicly available. The remaining ones are neither accessible nor traceable to specific data providers and a large fraction of these data might have been lost entirely. The present paper is an effort to rescue some of the data collected over the years following the CNPP accident and make them publicly available. The database includes surface air activity concentrations and deposition observations for (131)I, (134)Cs and (137)Cs measured and provided by Former Soviet Union authorities the years that followed the accident. Using the same interpolation tool as the official authorities, we have reconstructed a deposition map of (137)Cs based on about 3% of the data used to create the Atlas map. The reconstructed deposition map is very similar to the official one, but it has the advantage that it is based exclusively on documented data sources, which are all made available within this publication. In contrast to the official map, our deposition map is therefore reproducible and all underlying data can be used also for other purposes. The efficacy of the database was proved using simulated activity concentrations and deposition of (137)Cs from a Langrangian and a Euleurian transport model.