Collection Of Datasets Research Articles

Two years of volatile organic compound online in situ measurements at the Site Instrumental de Recherche par Télédétection Atmosphérique (Paris region, France) using proton-transfer-reaction mass spectrometry

Abstract. Volatile organic compounds (VOCs) have direct influences on air quality and climate. They indeed play a key role in atmospheric chemistry as precursors of secondary pollutants, such as ozone (O3) and secondary organic aerosols (SOA). In this respect, long-term datasets of in situ atmospheric measurements are crucial for characterizing the variability of atmospheric chemical composition, its sources, and trends. The ongoing establishment of the Aerosols, Cloud, and Trace gases Research InfraStructure (ACTRIS) allows implementation of the collection and provision of such high-quality datasets. In this context, online and continuous measurements of O3, nitrogen oxides (NOx), and aerosols have been carried out since 2012 at the SIRTA (Site Instrumental de Recherche par Télédétection Atmosphérique) observatory, located in the Paris region, France. Within the last decade, VOC measurements were conducted offline at SIRTA, until the implementation of real-time monitoring which started in January 2020 using a proton-transfer-reaction quadrupole mass spectrometer (PTR-Q-MS). The dataset acquired during the first 2 years of online VOC measurements provides insights into their seasonal and diurnal variabilities. The additional long-term datasets obtained from co-located measurements (NOx, aerosol physical and chemical properties, meteorological parameters) are used to better characterize the atmospheric conditions and to further interpret the obtained results. Results also include insights into VOC main sources and the influence of meteorological conditions and air mass origin on their levels in the Paris region. Due to the COVID-19 pandemic, the year 2020 notably saw a quasi-total lockdown in France in spring and a lighter one in autumn. Therefore, the focus is placed on the impact of these lockdowns on the VOC variability and sources. A change in the behaviour of VOC markers for anthropogenic sources was observed during the first lockdown, reflecting a change in human activities. A comparison with gas chromatography data from the Paris city centre consolidates the regional representativity of the SIRTA station for benzene, while differences are observed for shorter-lived compounds with a notable impact of their local sources. This dataset could be further used as input for atmospheric models and can be found at https://doi.org/10.14768/f8c46735-e6c3-45e2-8f6f-26c6d67c4723 (Simon et al., 2022a).

Early detection of Solanum lycopersicum diseases from temporally-aggregated hyperspectral measurements using machine learning

Some plant diseases can significantly reduce harvest, but their early detection in cultivation may prevent those consequential losses. Conventional methods of diagnosing plant diseases are based on visual observation of crops, but the symptoms of various diseases may be similar. It increases the difficulty of this task even for an experienced farmer and requires detailed examination based on invasive methods conducted in laboratory settings by qualified personnel. Therefore, modern agronomy requires the development of non-destructive crop diagnosis methods to accelerate the process of detecting plant infections with various pathogens. This research pathway is followed in this paper, and an approach for classifying selected Solanum lycopersicum diseases (anthracnose, bacterial speck, early blight, late blight and septoria leaf) from hyperspectral data captured on consecutive days post inoculation (DPI) is presented. The objective of that approach was to develop a technique for detecting infection in less than seven days after inoculation. The dataset used in this study included hyperspectral measurements of plants of two cultivars of S. lycopersicum: Benito and Polfast, which were infected with five different pathogens. Hyperspectral reflectance measurements were performed using a high-spectral-resolution field spectroradiometer (350–2500 nm range) and they were acquired for 63 days after inoculation, with particular emphasis put on the first 17 day-by-day measurements. Due to a significant data imbalance and low representation of measurements on some days, the collective datasets were elaborated by combining measurements from several days. The experimental results showed that machine learning techniques can offer accurate classification, and they indicated the practical utility of our approaches.

Cross-Study Replicability in Cluster Analysis.

In cancer research, clustering techniques are widely used for exploratory analyses, playing a critical role in the identification of novel cancer subtypes and patient management. As data collected by multiple research groups grows, it is increasingly feasible to investigate the replicability of clustering procedures, that is, their ability to consistently recover biologically meaningful clusters across several data sets. In this paper, we review methods for replicability of clustering analyses, and discuss a novel framework for evaluating cross-study clustering replicability, useful when two or more studies are available. Our approach can be applied to any clustering algorithm and can employ different measures of similarity between partitions to quantify replicability, globally (i.e., for the whole sample) as well as locally (i.e., for individual clusters). Using experiments on synthetic and real gene expression data, we illustrate the usefulness of our procedure to evaluate if the same clusters are identified consistently across a collection of data sets.

Genomic benchmarks: a collection of datasets for genomic sequence classification

BackgroundRecently, deep neural networks have been successfully applied in many biological fields. In 2020, a deep learning model AlphaFold won the protein folding competition with predicted structures within the error tolerance of experimental methods. However, this solution to the most prominent bioinformatic challenge of the past 50 years has been possible only thanks to a carefully curated benchmark of experimentally predicted protein structures. In Genomics, we have similar challenges (annotation of genomes and identification of functional elements) but currently, we lack benchmarks similar to protein folding competition.ResultsHere we present a collection of curated and easily accessible sequence classification datasets in the field of genomics. The proposed collection is based on a combination of novel datasets constructed from the mining of publicly available databases and existing datasets obtained from published articles. The collection currently contains nine datasets that focus on regulatory elements (promoters, enhancers, open chromatin region) from three model organisms: human, mouse, and roundworm. A simple convolution neural network is also included in a repository and can be used as a baseline model. Benchmarks and the baseline model are distributed as the Python package ‘genomic-benchmarks’, and the code is available at https://github.com/ML-Bioinfo-CEITEC/genomic_benchmarks.ConclusionsDeep learning techniques revolutionized many biological fields but mainly thanks to the carefully curated benchmarks. For the field of Genomics, we propose a collection of benchmark datasets for the classification of genomic sequences with an interface for the most commonly used deep learning libraries, implementation of the simple neural network and a training framework that can be used as a starting point for future research. The main aim of this effort is to create a repository for shared datasets that will make machine learning for genomics more comparable and reproducible while reducing the overhead of researchers who want to enter the field, leading to healthy competition and new discoveries.

Iron deficiency anemia detection using machine learning models: A comparative study of fingernails, palm and conjunctiva of the eye images

AbstractAnemia is one of the global public health challenges that particularly affect children and pregnant women. A study by WHO indicates that 42% of children below the age of 6 and 40% of pregnant women worldwide are anemic. This affects the world's total population by 33%, due to the cause of iron deficiency. The non‐invasive technique, such as the use of machine learning algorithms is one of the methods used in the diagnosis or detection of clinical diseases, which anemia detection cannot be overlooked in recent days. In this study, a machine learning approach was used to detect iron‐deficiency anemia with the application of Naïve Bayes, CNN, SVM, k‐NN, and decision tree algorithms. This enabled us to compare the conjunctiva of the eyes, the palpable palm, and the color of the fingernail images to justify which of them has a higher accuracy for detecting anemia in children. The method utilized was categorized into three different stages: dataset collection, dataset preprocessing, and model development for anemia detection. The CNN achieved a higher accuracy of 99.12%, while the SVM had the least accuracy of 95.4%. The performance of the models justifies that the non‐invasive approach is an effective mechanism for anemia detection.

163 Assessment of implementation, needs, and challenges of equine Extension programs

Extension programs aim to provide quality education and solutions to problems in agriculture through the use of research-based information. While youth and adult equine Extension programs are available for implementation at the county, regional, and state level, equine-specific programs are not equally utilized or adopted compared with other livestock programs. The objective of this study was to evaluate how equine Extension programs are utilized and delivered in TN counties and to understand perceived barriers for program implementation by Extension personnel. An equine program needs assessment survey (Qualtrics, Provo, UT) was distributed to all Extension County Directors, Agriculture & Natural Resources (ANR) and 4-H Agents and Specialists in TN from February to May 2022. Respondents were prompted with programming questions including demographics, active programming, types of equine programs offered, resources needed, and perceived barriers to implementing programs. Based on removal of individual identifiers from this post collection data set, the University of TN Institutional Review Board deemed this work as non-human subject research and subsequent approval was not required. Descriptive statistics were calculated using Stata/SE 17 (College Station, TX). A total of 44 counties and regions were represented, where the highest proportion of respondent (n = 61) positionappointments were ANR County Agents (49.18%, n = 30), 4-H County Agents (34.43%, n = 21), and split appointment ANR and 4-H County Agents (6.56%, n = 4). Currently, active equine programming at the county level (n = 55) is mostly youth focused (52.73%, n = 29), and many counties currently do not have an active program (36.36%, n = 20). Respondents (n = 61) indicated the largest need for Extension publications (68.85%, n = 42), how-to videos (57.38%, n = 35), and news releases (57.38%, n = 35) for use at the county level. When specifically asked why developed programs were not adopted at the county level, respondents (n = 60) indicated time constraints (21.67%, n = 13), lack of interest (21.67%, n = 13), and cost (21.67%, n = 13) as the largest perceived barriers. Specific to cost, respondents (n = 61) felt program costs should include a manual and educational materials (55.74%, n = 34), program sign (49.18%, n = 30), and a state processing fee (44.26%, n = 27) when registration or program coordination is handled outside of the county. Equine-specific Extension programs do not appear to be well-adopted at the county level in Tennessee for various reasons. Creative efforts must be made to implement equine-specific Extension programs and meet equine clientele needs.

Benefits of using blended generative adversarial network images to augment classification model training data sets

Object detection algorithms have reached nearly superhuman levels within the last decade; however, these algorithms require large diverse training data sets to ensure their operational performance matches performance demonstrated during testing. The collection and human labeling of such data sets can be expensive and, in some cases, such as Intelligence, Surveillance and Reconnaissance of rare events it may not even be feasible. This research proposes a novel method for creating additional variability within the training data set by utilizing multiple models of generative adversarial networks producing both high- and low-quality synthetic images of vehicles and inserting those images alongside images of real vehicles into real backgrounds. This research demonstrates a 17.90% increase in mean absolute percentage error, on average, compared to the YOLOv4-Tiny Model trained on the original non-augmented training set as well as a 14.44% average improvement in the average intersection over union rate. In addition, our research adds to a small, but growing, body of literature indicating that the inclusion of low-quality images into training data sets is beneficial to the performance of computer vision models.

From market to environment – consumption-normalised pharmaceutical emissions in the Rhine catchment

Direct and indirect threats by organic micropollutants can only be reliably assessed and prevented if the exposure to these chemicals is known, which in turn requires a confident estimate of their emitted amounts into the environment. APIs (Active Pharmaceutical Ingredients) enter surface waters mostly through the sewer system and wastewater treatment plants (WWTPs). However, their effluent fluxes are highly variable and influenced by several different factors that challenge robust emission estimates. Here, we defined a dimensionless, theoretically consumption-independent ‘escape factor’ (kesc) for estimating the amount of APIs (expected to be) present in WWTP effluents. The factor is determined as the proportion of marketed and actually emitted amounts of APIs. A large collection of German and Swiss monitoring datasets were analyzed to calculate stochastic kesc values for 31 APIs, reflecting both the magnitude and uncertainty of consumption-normalised emissions. Escape factors provide an easy-to-use tool for the estimation of average API emissions and expected variability from numerous WWTPs given that consumption data are provided, thereby supporting simulation modeling of the fate of APIs in stream networks or exposure assessments.

Impact of Fireworks Industry Safety Measures and Prevention Management System on Human Error Mitigation Using a Machine Learning Approach.

In the fireworks industry (FI), many accidents and explosions frequently happen due to human error (HE). Human factors (HFs) always play a dynamic role in the incidence of accidents in workplace environments. Preventing HE is a main challenge for safety and precautions in the FI. Clarifying the relationship between HFs can help in identifying the correlation between unsafe behaviors and influential factors in hazardous chemical warehouse accidents. This paper aims to investigate the impact of HFs that contribute to HE, which has caused FI disasters, explosions, and incidents in the past. This paper investigates why and how HEs contribute to the most severe accidents that occur while storing and using hazardous chemicals. The impact of fireworks and match industry disasters has motivated the planning of mitigation in this proposal. This analysis used machine learning (ML) and recommends an expert system (ES). There were many significant correlations between individual behaviors and the chance of HE to occur. This paper proposes an ML-based prediction model for fireworks and match work industries in Sivakasi, Tamil Nadu. For this study analysis, the questionnaire responses are reviewed for accuracy and coded from 500 participants from the fireworks and match industries in Tamil Nadu who were chosen to fill out a questionnaire. The Chief Inspectorate of Factories in Chennai and the Training Centre for Industrial Safety and Health in Sivakasi, Tamil Nadu, India, significantly contributed to the collection of accident datasets for the FI in Tamil Nadu, India. The data are analyzed and presented in the following categories based on this study's objectives: the effect of physical, psychological, and organizational factors. The output implemented by comparing ML models, support vector machine (SVM), random forest (RF), and Naïve Bayes (NB) accuracy is 86.45%, 91.6%, and 92.1%, respectively. Extreme Gradient Boosting (XGBoost) has the optimal classification accuracy of 94.41% of ML models. This research aims to create a new ES to mitigate HE risks in the fireworks and match work industries. The proposed ES reduces HE risk and improves workplace safety in unsafe, uncertain workplaces. Proper safety management systems (SMS) can prevent deaths and injuries such as fires and explosions.

A Literature Survey on AI-Aided Beamforming and Beam Management for 5G and 6G Systems.

Modern wireless communication systems rely heavily on multiple antennas and their corresponding signal processing to achieve optimal performance. As 5G and 6G networks emerge, beamforming and beam management become increasingly complex due to factors such as user mobility, a higher number of antennas, and the adoption of elevated frequencies. Artificial intelligence, specifically machine learning, offers a valuable solution to mitigate this complexity and minimize the overhead associated with beam management and selection, all while maintaining system performance. Despite growing interest in AI-assisted beamforming, beam management, and selection, a comprehensive collection of datasets and benchmarks remains scarce. Furthermore, identifying the most-suitable algorithm for a given scenario remains an open question. This article aimed to provide an exhaustive survey of the subject, highlighting unresolved issues and potential directions for future developments. The discussion encompasses the architectural and signal processing aspects of contemporary beamforming, beam management, and selection. In addition, the article examines various communication challenges and their respective solutions, considering approaches such as centralized/decentralized, supervised/unsupervised, semi-supervised, active, federated, and reinforcement learning.

Recent Advances in NMR Protein Structure Prediction with ROSETTA.

Nuclear magnetic resonance (NMR) spectroscopy is a powerful method for studying the structure and dynamics of proteins in their native state. For high-resolution NMR structure determination, the collection of a rich restraint dataset is necessary. This can be difficult to achieve for proteins with high molecular weight or a complex architecture. Computational modeling techniques can complement sparse NMR datasets (<1 restraint per residue) with additional structural information to elucidate protein structures in these difficult cases. The Rosetta software for protein structure modeling and design is used by structural biologists for structure determination tasks in which limited experimental data is available. This review gives an overview of the computational protocols available in the Rosetta framework for modeling protein structures from NMR data. We explain the computational algorithms used for the integration of different NMR data types in Rosetta. We also highlight new developments, including modeling tools for data from paramagnetic NMR and hydrogen-deuterium exchange, as well as chemical shifts in CS-Rosetta. Furthermore, strategies are discussed to complement and improve structure predictions made by the current state-of-the-art AlphaFold2 program using NMR-guided Rosetta modeling.

ACVECC‐Veterinary Committee on Trauma registry report 2017–2019

To report summative data from the American College of Veterinary Emergency and Critical Care Veterinary Committee on Trauma (VetCOT) registry. Multi-institutional registry data report, April 1, 2017 to December 31, 2019. VetCOT identified and verified Veterinary Trauma Centers (VTCs). Dogs and cats with evidence of trauma. Data were input to a web-based data capture system (Research Electronic Data Capture) by data entry personnel trained in data software use and operational definitions of data variables. Data on demographics, trauma type, preadmission care, trauma severity assessment at presentation (modified Glasgow Coma Scale and Animal Trauma Triage score), key laboratory parameters, interventions, and case outcome were collected. Summary descriptive data for each species are reported. Thirty-one VTCs contributed data from 20,842 canine and 4003 feline trauma cases during the 33-month reporting period. Most cases presented directly to a VTC (82.1% dogs, 82.1% cats). Admission to hospital rates were slightly lower in dogs (27.8%) than cats (32.7%). Highest mortality rates by mechanism of injury in dogs were struck by vehicle (18.3%), ballistic injury (17.6%), injured inside vehicle (13.2%), nonpenetrating bite wound (10.2%), and choking/pulling injury (8.5%). Highest mortality rates by mechanism of injury in cats were struck by vehicle (43.3%), ejected from vehicle (33.3%), nonpenetrating bite wound (30.7%), ballistic injury (27.8%), and choking/pulling injury (25.0%). The proportion of animals surviving to discharge was 93.1% (dogs) and 82.5% (cats). The VetCOT registry is a powerful resource for collection of a large dataset on trauma in dogs and cats seen at VTCs. Overall survival to discharge was high indicating low injury severity for most recorded cases. Further evaluation of data on subsets of injury types, patient assessment parameters, interventions, and associated outcome are warranted. Data from the registry can be leveraged to inform clinical trial design and justification for naturally occurring trauma as a translational model to improve veterinary and human trauma patient outcome.

Comparison of performances of ML-Algorithms in the estimation of the execution time of non-parallel Java programs

Performance enhancement of a computer program is an important aspect of today's world. Developers produce programs, and there is a lack of accurate methods for predicting the execution time of a computer program prior to its execution in an executable environment. Predicting the execution time of a particular program before execution would be great for developing the program with the highest performance efficiency and with the lowest execution latency. This research introduces a Machine Learning based solution to predict an execution-time-based label for a given computer program. There are three main types of parameters in a computer program that affect the execution time, such as Static Code Features, HTTP Calls, and the Hardware Performance of the execution environment. In this research, the Machine Learning model was trained for the parameters of the above types (Programs with Static Programs &amp; HTTP calls) by executing them on a fixed hardware infrastructure execution condition. We analyzed the number of if conditions, methods, breaks, switches, loops, nested-loop-depth, frequencies, and the behaviour of HTTP calls, kind of features of a computer program in order to generate an accurate execution time complexity prediction label of a computer program. Further, in the collected data set, the most prominent feature which affects the complexity among the features that we considered is the number of HTTP calls and nested loop depth, followed by loops. Accuracy Score, Precision, Recall, and F1 Score values of the ML model were generated for the traditional classification algorithms such as Decision Tree Classifier, K Nearest Neighbor Classifier, Random Forest Classifier, Naive Bayes Classifier, Support Vector Classifier, and MLP Classifiers in order to verify the effectiveness of the model. The best accuracy score was achieved with an overall 88% by using the approach of Random Forest. The findings of this research can be optimized for implementing an IDE plugin or a developer tool that can predict the exact execution time of a given computer program live by integrating the specifications of the execution device. It will help developers in terms of optimizing a particular computer program and developing it for a minimum execution latency and enhancing the performance of the program.

Abstract LB073: NCI’s broad-use collections: Accelerating discovery process by improving access to individual-level genomics and other -omics data

Abstract Genomic data sharing increases our understanding of factors that influence health and diseases by enabling additional research questions from secondary data users, increasing statistical power through combining multiple data sources, facilitating reproducibility and validation of research results, and supporting innovation with the development of research tools and methodologies. Research using large-scale genomic and other-omics data is no longer limited by storage space or slow downloading speed for individual datasets. Innovations like cloud infrastructure enable computing over many datasets at multiple locations at once. These developments have increased the need for faster, more efficient processes for data access and sharing. To support scientific exploration and meet the demand for analysis of biomedical data, the National Cancer Institute (NCI) has created two large collections of the broad-use studies within the database of Genotypes and Phenotypes (dbGaP). These collections comply with the consent of the study (data use limitations) for how secondary access to studies are determined. NCI’s Collection of Datasets for General Research Use comprises 284 studies with individual-level data sets, and permits approved users to explore broad research interest, including methods and tool development. NCI’s Collection of Datasets for Health, Medical, and Biomedical (HMB) Research Purposes is comprised of 65 studies of individual-level that are permitted for research interests specific to any health, medical, or biomedical research only. The HMB collection could be used for methods and tool development; research interests involving ancestry/populations studies must be dependent on a health/medical condition. Through these collections, investigators will have the potential to add access to 349 datasets to their approved research projects or submit a new project request for access to the broad-use collections. As new genomic studies are registered by NCI for release through dbGaP, they will be automatically added to these collections. NCI anticipates through the implementation of these broad-use collections a requestor could have access to approximately 70% of NCI’s studies in controlled-access repositories.The streamlined access to broad-use datasets expedites data sharing and potentially accelerates the discovery process. This approach reduces redundancies for obtaining controlled-access data. Citation Format: Freddie L. Pruitt, Michael Feolo, Subhashini Jagu, Jaime Guidry Auvil. NCI’s broad-use collections: Accelerating discovery process by improving access to individual-level genomics and other -omics data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB073.

Effect of selection bias on Automatic Colonoscopy Polyp Detection

Despite the successful demonstration of the deep learning architectures for Automatic Colonoscopy Polyp Detection (ACPD), studies have argued upon human bias, and lack of interpretability in the existing architectures which hampers its routine clinical and real-time use. To investigate the presence of selection bias, the present work develops a proof-of-concept model with the help of reverse psychology. The model comprises a fine-tuned, interpretable ACPD architecture (VGG-16) and its real-time application (app) called the polyp testing app (link) without consideration of selection bias in the existing open-source colonoscopy datasets. To check the robustness of the developed architecture, it has been trained, validated, and tested on six independent datasets. Evaluation methods in the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) criteria have been used to check the effectiveness of the architecture. Specificity and precision up to 97.3% and 92.0% respectively have been achieved on the validation dataset. A new test set called ‘Gastrolab-Polyp’ has been developed for fair external testing of the developed architecture (link). In comparison to the reference app, the developed app achieved a percentage improvement of 15%, 30.5%, and 14.5% in an overall test dataset accuracy, precision, and recall respectively for the developed Gastrolab-Polyp test dataset. Interpretability methods such as class activation maps, Shapley plots, LIME, and detection analysis of the developed and reference app indicate a bias towards polyp frames. The present work concludes that there is a selection bias of non-polyp frames in five open-source colonoscopy datasets used for polyp detection. Careful dataset collection is needed for its use.

In-TFK: a scalable traditional food knowledge platform, a new traditional food dataset, platform, and multiprocess inference service

Traditional Food Knowledge (TFK) is needed to define the acculturation of culture, society, and health in the context of food. TFK is essential for a human’s cultural, economic, and health aspects. Variations of ethnicity, culture, and lifestyle affect the diversity of traditional Indonesian food. Recognition of food is needed to maintain the sustainability of traditional food. Nowadays, there are many food dataset collections, but there needs to be a dataset that specifically collects standard food datasets. Our main contributions to the TFK research field are professional food image data acquisition, innovative development of an automatic, scalable food recognition system, and multi-process inference service. There are 34 variations of traditional foods from all regions in Indonesia that were acquired in this dataset. The dataset comprises 1644 high-quality images captured by professional cameras and 1020 by a smartphone. Several deep learning models are implemented in food recognition systems. This system can accommodate the addition and reduction of food variations in the knowledge recognition system and is capable of multiple concurrent requests at a time. The current prototype incorporates traditional types of food from Indonesia. However, the food model can also be expanded to other countries traditional foods. The automatic recognition systems are evaluated using several deep-learning network models. The experiment results have shown that the AUROC score is 0.99, and the request success rate can be improved by 70% with a multiprocess inference service.

Fast-Training Deep Learning Algorithm for Multiplex Quantification of Mammalian Bioproduction Metabolites via Contactless Short-Wave Infrared Hyperspectral Sensing.

Within the biopharmaceutical sector, there exists the need for a contactless multiplex sensor, which can accurately detect metabolite levels in real time for precise feedback control of a bioreactor environment. Reported spectral sensors in the literature only work when fully submerged in the bioreactor and are subject to probe fouling due to a cell debris buildup. The use of a short-wave infrared (SWIR) hyperspectral (HS) cam era allows for efficient, fully contactless collection of large spectral datasets for metabolite quantification. Here, we report the development of an interpretable deep learning system, a convolution metabolite regression (CMR) approach that detects glucose and lactate concentrations using label-free contactless HS images of cell-free spent media samples from Chinese hamster ovary (CHO) cell growth flasks. Using a dataset of <500 HS images, these CMR algorithms achieved a competitive test root-mean-square error (RMSE) performance of glucose quantification within 27 mg/dL and lactate quantification within 20 mg/dL. Conventional Raman spectroscopy probes report a validation performance of 26 and 18 mg/dL for glucose and lactate, respectively. The CMR system trains within 10 epochs and uses a convolution encoder with a sparse bottleneck regression layer to pick the best-performing filters learned by CMR. Each of these filters is combined with existing interpretable models to produce a metabolite sensing system that automatically removes spurious predictions. Collectively, this work will advance the safe and efficient adoption of contactless deep learning sensing systems for fine control of a variety of bioreactor environments.

Foundation models for generalist medical artificial intelligence.

The exceptionally rapid development of highly flexible, reusable artificial intelligence (AI) models is likely to usher in newfound capabilities in medicine. We propose a new paradigm for medical AI, which we refer to as generalist medical AI (GMAI). GMAI models will be capable of carrying out a diverse set of tasks using very little or no task-specific labelled data. Built through self-supervision on large, diverse datasets, GMAI will flexibly interpret different combinations of medical modalities, including data from imaging, electronic health records, laboratory results, genomics, graphs or medical text. Models will in turn produce expressive outputs such as free-text explanations, spoken recommendations or image annotations that demonstrate advanced medical reasoning abilities. Here we identify a set of high-impact potential applications for GMAI and lay out specific technical capabilities and training datasets necessary to enable them. We expect that GMAI-enabled applications will challenge current strategies for regulating and validating AI devices for medicine and will shift practices associated with the collection of large medical datasets.

Enhancing Spam Message Classification and Detection Using Transformer-Based Embedding and Ensemble Learning.

Over the last decade, the Short Message Service (SMS) has become a primary communication channel. Nevertheless, its popularity has also given rise to the so-called SMS spam. These messages, i.e., spam, are annoying and potentially malicious by exposing SMS users to credential theft and data loss. To mitigate this persistent threat, we propose a new model for SMS spam detection based on pre-trained Transformers and Ensemble Learning. The proposed model uses a text embedding technique that builds on the recent advancements of the GPT-3 Transformer. This technique provides a high-quality representation that can improve detection results. In addition, we used an Ensemble Learning method where four machine learning models were grouped into one model that performed significantly better than its separate constituent parts. The experimental evaluation of the model was performed using the SMS Spam Collection Dataset. The obtained results showed a state-of-the-art performance that exceeded all previous works with an accuracy that reached 99.91%.

Julia for biologists.

Major computational challenges exist in relation to the collection, curation, processing and analysis of large genomic and imaging datasets, as well as the simulation of larger and more realistic models in systems biology. Here we discuss how a relative newcomer among programming languages-Julia-is poised to meet the current and emerging demands in the computational biosciences and beyond. Speed, flexibility, a thriving package ecosystem and readability are major factors that make high-performance computing and data analysis available to an unprecedented degree. We highlight how Julia's design is already enabling new ways of analyzing biological data and systems, and we provide a list of resources that can facilitate the transition into Julian computing.