Detailed Categories Research Articles

Intra-urban land use maps for a global sample of cities from Sentinel-2 satellite imagery and computer vision

Intra-urban land use maps provide important information to urban planners and policymakers, but these maps are costly, time consuming to generate and are often unavailable in developing countries where most urban growth is now occurring. This paper reports on machine learning methods to automate the production of land use maps from cloud-free mosaics of Sentinel-2 imagery. We have trained a novel neural network architecture to produced 5 meter resolution land use maps for a global stratified sample of 200 cities. The sample includes all world regions, 78 countries, and a range of population sizes. The model architecture is roughly 1 to 2 orders of magnitude smaller than similar architectures such as UNet (Ronneberger, Fischer, & Brox, 2015) and DeeplabV3+ (Chen, Zhu, Papandreou, Schroff, & Adam, 2018), significantly lowering the cost and computational requirements to produce maps. We are in the process of generating land use maps for all 4,000 + cities and metropolitan areas in the world with populations exceeding 100,000. The resulting product will be the first, regularly updated, freely available, global intra-urban land use maps at 5 meter resolution.We present a 4-tier land use taxonomy which at its root distinguishes open-space from built-up area. At the second tier, it subdivides the built-up category into nonresidential and residential areas. The third tier distinguishes formal from informal residential land use, and the fourth tier further subdivides formal and informal residential land uses into more detailed categories. Accuracy scores at tier-1 and tier-2 were 86% and 79% respectively. Tiers 3 and 4 had an accuracy scores of 75% and 71% respectively. Additionally, we train a roads-only model and compare its output to the Atlas of Urban Expansion’s Arterial Roads dataset and Open Street Map. As an example use case, we train an Informal Settlement Classifier, correctly classifying 87% of the settlements.

COF-Based Composites: Extraordinary Removal Performance for Heavy Metals and Radionuclides from Aqueous Solutions

Covalent-organic-framework-based composites (COF-based composites) are novel functional materials constructed by combining COFs and various organic/inorganic materials, which usually exhibit more advantageous physicochemical properties, stability, selectivity, and pollutant removal performance compared with pristine COFs. In this review, we highlighted the detailed categories and the recent advances of binary and ternary COF-based composites. In particular, the progress of COF-based composites for the efficient removal of heavy metals and radionuclides is comprehensively summarized. By comparing the performance with other materials, we found that the functional materials make the composites stand out. In addition, the technologies to investigate the internal mechanism of pollutants removed by COF-based composites, mainly including characterization analysis and theoretical calculations, are discussed and summarized, providing the direction for in-depth understanding of COF-based composites. According to the discussion above, the opportunities and challenges of COF-based composites in the treatment of heavy metals and radionuclides are summarized and prospected. Through this review, we desire to raise the awareness of COF-based composites and advance further studies in this field by proposing the potential research directions and applications.Graphical Abstract

Confidentiality Protection in the 2020 US Census of Population and Housing

In an era where external data and computational capabilities far exceed statistical agencies’ own resources and capabilities, they face the renewed challenge of protecting the confidentiality of underlying microdata when publishing statistics in very granular form and ensuring that these granular data are used for statistical purposes only. Conventional statistical disclosure limitation methods are too fragile to address this new challenge. This article discusses the deployment of a differential privacy framework for the 2020 US Census that was customized to protect confidentiality, particularly the most detailed geographic and demographic categories, and deliver controlled accuracy across the full geographic hierarchy.

Remote sensing sea ice classification based on DenseNet and heterogeneous data fusion

Most remote sensing sea ice classification methods use single-source remote sensing data, such as synthetic aperture radar (SAR) data and optical remote sensing data. SAR data contain rich sea ice texture information, but the data are relatively single, making it difficult to distinguish detailed sea ice categories. Optical data include abundant spatial-spectral information, but they are often affected by clouds, fog, and severe weather. Hence, given the limitations of single-source data, the remote sensing sea ice classification accuracy cannot be further improved. A sea ice classification method based on deep learning and multisource remote sensing data fusion is proposed utilizing an improved densely connected convolutional neural network (DenseNet) to mine and fuse the multilevel features of sea ice. According to the characteristics of SAR data and optical data, a dual-branch network structure based on an improved DenseNet is employed for feature extraction, and a squeeze-and-excitation attention mechanism is introduced to weight the fusion features to further enhance the feature weights that can effectively distinguish different types of sea ice. The fully connected network is used to perform the deep fusion of features and classify sea ice. To verify the effectiveness of the proposed method, two sets of sea ice data are utilized for sea ice classification. The experimental results show that the proposed method fully excavates and fuses the multilevel characteristics of heterogeneous data using the improved dual-branch network structure, leverages the complementary characteristics of SAR data and optical data, significantly increases the sea ice classification accuracy, and effectively improves the influence of cloud cover on the sea ice classification accuracy. Compared with the typical single-source data classification method and other heterogeneous data fusion classification methods, the proposed method achieves superior overall classification accuracy (98.49% and 98.58%).

Land Use/Land Cover Mapping Based on GEE for the Monitoring of Changes in Ecosystem Types in the Upper Yellow River Basin over the Tibetan Plateau

The upper Yellow River basin over the Tibetan Plateau (TP) is an important ecological barrier in northwestern China. Effective LULC products that enable the monitoring of changes in regional ecosystem types are of great importance for their environmental protection and macro-control. Here, we combined an 18-class LULC classification scheme based on ecosystem types with Sentinel-2 imagery, the Google Earth Engine (GEE) platform, and the random forest method to present new LULC products with a spatial resolution of 10 m in 2018 and 2020 for the upper Yellow River Basin over the TP and conducted monitoring of changes in ecosystem types. The results indicated that: (1) In 2018 and 2020, the overall accuracy (OA) of LULC maps ranged between 87.45% and 93.02%. (2) Grassland was the main LULC first-degree class in the research area, followed by wetland and water bodies and barren land. For the LULC second-degree class, the main LULC was grassland, followed by broadleaf shrub and marsh. (3) In the first-degree class of changes in ecosystem types, the largest area of progressive succession (positive) was grassland–shrubland (451.13 km2), whereas the largest area of retrogressive succession (negative) was grassland–barren (395.91 km2). In the second-degree class, the largest areas of progressive succession (positive) were grassland–broadleaf shrub (344.68 km2) and desert land–grassland (302.02 km2), whereas the largest areas of retrogressive succession (negative) were broadleaf shrubland–grassland (309.08 km2) and grassland–bare rock (193.89 km2). The northern and southwestern parts of the study area showed a trend towards positive succession, whereas the south-central Huangnan, northeastern Gannan, and central Aba Prefectures showed signs of retrogressive succession in their changes in ecosystem types. The purpose of this study was to provide basis data for basin-scale ecosystem monitoring and analysis with more detailed categories and reliable accuracy.

Teacher Candidates’ Self-Determined Motivation to Develop and Implement Self-Regulated Learning Practices

Teaching towards self-regulated learning (SRL) is complex and involves the development of skills and sustained motivation. This study examined teacher candidates’ (TCs’) identification of supports and constraints for their self-determined motivation to develop SRL practices. Findings from one case within a qualitative, longitudinal study of four teacher candidates enrolled in a teacher education program (TEP) focused on SRL in Canada are presented. Supports and constraints for this TC’s self-determined motivation in relation to her development and implementation of self-regulated promoting practices are identified and discussed from the perspective of SRL and self-determination theory. The data analyzed included: a questionnaire, interviews, documents, and in-class observations. The finding reveals detailed descriptive codes and categories for SRL and management practices, as well as codes related to TCs’ motivational constraints and affordances for their development of SRL practices.

Obliquely Loaded Welded Attachments Fatigue Categorization in Steel Bridges

The detail category for base metal at the toe of transverse stiffener-to-flange and transverse stiffener-to-web fillet welds is defined as Category C' in the current AASHTO LRFD Bridge Design Specifications (2020, 9th Ed.) and as Category C in the AREMA (2020) Manual for Railway Engineering and the AISC Steel Construction Manual (2017, 15th Ed.). These are often referred to as short attachments due to their very short length (< 2 in.) in the direction of the primary stress range. Sometimes it is necessary to place a stiffener or a connection plate at an angle different than perpendicular to the web, such as in skewed bridges. Increases in the effective length of the stiffener along the flange in the longitudinal direction are seen as the plate is rotated away from being perpendicular to the web. The other extreme occurs when the stiffener is rotated completely 90° and is perfectly parallel to the web and the longitudinal stress range. In this instance, this is identical to the long attachment and classified as Category E (length > 4 in.). The current specifications and manuals, on the other hand, do not have classification on how to address the potential effects on fatigue performance of angles in between these two extremes. This paper summarizes finite element analysis studies based on local stress and structural hot-spot stress approaches that were conducted to investigate and classify welded attachments placed at angles other than 0° (transverse) or 90° (longitudinal) for a variety of stiffener geometries and thicknesses. This study includes new classification for incorporating the findings into the AASHTO LRFD Bridge Design Specifications, AREMA Manual for Railway Engineering, and the AISC Steel Construction Manual.

Progress of Nanomaterials-Based Photothermal Therapy for Oral Squamous Cell Carcinoma

Oral squamous cell carcinoma (OSCC) is one of the top 15 most prevalent cancers worldwide. However, the current treatment models for OSCC (e.g., surgery, chemotherapy, radiotherapy, and combination therapy) present several limitations: damage to adjacent healthy tissue, possible recurrence, low efficiency, and severe side effects. In this context, nanomaterial-based photothermal therapy (PTT) has attracted extensive research attention. This paper reviews the latest progress in the application of biological nanomaterials for PTT in OSCC. We divide photothermal nanomaterials into four categories (noble metal nanomaterials, carbon-based nanomaterials, metal compounds, and organic nanomaterials) and introduce each category in detail. We also mention in detail the drug delivery systems for PTT of OSCC and briefly summarize the applications of hydrogels, liposomes, and micelles. Finally, we note the challenges faced by the clinical application of PTT nanomaterials and the possibility of further improvement, providing direction for the future research of PTT in OSCC treatment.

Differences in COVID-19 vaccination coverage by occupation in England: a national linked data study

BackgroundMonitoring differences in COVID-19 vaccination uptake in different groups is crucial to help inform the policy response to the pandemic. A key data gap is the absence of data on...

Lifecycle Environmental Benefits with a Hybrid Electric Propulsion System Using a Control Algorithm for Fishing Boats in Korea

In Korea, fishing boats account for 65,452 out of 71,174 coastal ships, and aging ships are a leading cause of environmental problems and economic loss. To solve this problem, the government has enacted a policy to promote the development and distribution of environmental vessels. Electric propulsion systems and hybrid systems are suggested as alternatives. However, to study whether this hybrid system is environmentally friendly, a system for measuring the performance of conventional diesel-powered ships was developed and measured. A case study applying the battery hybrid system was analyzed through a Matlab/Simulink simulation, and the effect of reducing the carbon dioxide emissions by 7.6% was verified. In addition, comparing with a case where the load-sharing optimal control was introduced, we proved that the carbon dioxide reduction with the controller was 10.7%. By applying the results of this study to various countries with different fuel production processes through the LCA technique, it was confirmed that the battery hybrid system is eco-friendly from the energy production stage, rather than unconditionally, according to the national power production method. Additional research is needed to reflect the aging trend of the entire fleet and to create a detailed category classification through future research.

Auditory Sequences Presented With Spearcons Support Better Multiple Patient Monitoring Than Single-Patient Alarms: A Preclinical Simulation.

A study of auditory displays for simulated patient monitoring compared the effectiveness of two sound categories (alarm sounds indicating general risk categories from international alarm standard IEC 60601-1-8 versus event-specific sounds according to the type of nursing unit) and two configurations (single-patient alarms versus multi-patient sequences). Fieldwork in speciality-focused high dependency units (HDU) indicated that auditory alarms are ambiguous and do not identify which patient has a problem. We tested whether participants perform better using auditory displays that identify the relevant patient and problem. During simulated patient monitoring of four patients in a respiratory HDU, 60 non-clinicians heard either (a) IEC risk categories as single-patient alarm sounds, (b) event-specific categories as single-patient alarm sounds, (c) IEC risk categories in multi-patient sequences or (d) event-specific categories in multi-patient sequences. Participants performed a perceptual-motor task while monitoring patients; after detecting abnormal events, they identified the patient and the event. Participants hearing multi-patient sequences made fewer wrong patient identifications than participants hearing single-patient alarms. Advantages of event-specific categories emerged when IEC risk category sounds indicated more than one potential event. Even when IEC and event-specific sounds indicated the same unique event, spearcons supported better event identification than did auditory icon sounds. Auditory displays that unambiguously convey which patient is having what problem dramatically improve monitoring performance in a preclinical HDU simulation. Time-compressed speech assists development of detailed risk categories needed in specific HDU contexts, and multi-patient sound sequences allow multiple patient wellbeing to be monitored.

Malware and their diverse characteristics related to detection and analysis: A literature survey

The term malware refers to a specific form of software that causes damage to the computing device through data leakage and transformation, device malfunctioning, hacking, and exploitation. Typical malware can be categorized into several categories including viruses, worms, trojans, ransomware, etc. The analysis of malware helps in the categorization and behavior judgment of malware. The deep analysis of malware helps in malware detection. The current research study covers a hierarchical representation of various malware categories, different forms of detection, computing devices for malware, and a malware analysis approach. The study describes each category in detail, which helps in forming the mapping of malware analysis in detail.

COVID‐19 test positivity by occupation using the Delphi US COVID‐19 trends and impact survey, September–November 2020

BackgroundThe potential for work to be a risk factor for coronavirus disease 2019 (COVID‐19) was recognized early in the pandemic based on the likelihood of work‐related differences in exposures to COVID‐19 in different occupations. Due to intense demands of the pandemic, implementation of recommendations to collect information on occupation in relation to COVID‐19 has been uneven across the United States. The objective of this study was to investigate COVID‐19 test positivity by occupation.MethodsWe analyzed data collected from September 8 to November 30, 2020, by the Delphi Group at Carnegie Mellon University US COVID‐19 Trends and Impact Survey, offered daily to a random sample of US‐based Facebook users aged 18 years or older, who were invited via a banner in their news feed. Our focus was ever testing positive for COVID‐19 in respondents working outside the home for pay in the past 4 weeks.ResultsThe major occupational groups of “Production", "Building and grounds cleaning and maintenance,” “Construction and extraction,” “Healthcare support,” and “Food preparation and serving” had the five highest test positivity percentages (16.7%–14.4%). Highest detailed occupational categories (28.6%–19.1%) were “Massage therapist,” “Food processing worker,” “Bailiff, correctional officer, or jailer,” “Funeral service worker,” “First‐line supervisor of production and operating workers,” and “Nursing assistant or psychiatric aide.” Differences in test positivity by occupation remained after adjustment for age, gender, and pre‐existing medical conditions.ConclusionInformation on differences in test positivity by occupation can aid targeting of messaging for vaccination and testing and mitigation strategies for the current and future respiratory infection epidemics and pandemics. These results, obtained before availability of COVID‐19 vaccines, can form a basis for comparison to evaluate impacts of vaccination and subsequent emergence of viral variants.

How Do Non-professional Participants of a Trial Cope with the Communication Process at the Trial? The Results of Empirical Research Conducted in Polish Courts.

The aim of this article is to present some of the results of empirical research on the communication process at a trial conducted in Polish courts. These results will concern the participation of non-professional participants of a trial and the ways in which they deal with the communication process in the courtroom. The article presents the results of the analysis of the research material conducted in accordance with the detailed research questions and analytical categories. The analysis has especially shown that: (1) the non-professional participants used some legal terminology, but the statements without legal terminology were also communicatively effective; (2) the level of activity of non-professional participants related to participation at the trial varied depending on what the activity concerned; the activity in asking questions to the presiding judges regarding legal issues and the course of the court proceedings appeared to be significant; (3) the non-professional participants’ had a real and significant problem with asking questions during the examination (the proper realization of this element of the trial); (4) the statements of the non-professional participants of a trial were very protective; they used many different “linguistic means of protection” (e.g., acts of supposition or acts of doubt) which are connected with the obligation to tell the truth, the prohibition of concealing the truth and giving false testimony during examination; (5) the non-professional participants of a trial reach for adequate argumentative acts, which are suitable for influencing the perception of specific events or persons, and they strengthen their statements by using appropriate linguistic means of persuasion.

On the impact of undercuts and lack of penetration of cruciform joints from fatigue tests

The assumption of the detail categories in EN1993-1-9 is the presence of a certain weld quality level. In order to find out the sensitivity of the resulting fatigue strength for cases where a worse or better quality level compared to the selected level of EN ISO 5817 is achieved, in this paper, special fatigue experiments are presented. The experiments refer to the load-carrying cruciform joint with lack of penetration and root fusion as well as undercuts. The performed fatigue investigations on cruciform joints are based on and accompanied by further fatigue test results of literature, documented in a detailed fatigue database containing 147 test series. Detailed information on mechanical properties, fabrication and metallurgical characteristics of the cruciform specimens are given. Extensive 3D laser scans of the specimens and ultrasonic testing as well as magnetic particle testing have been performed in order to analyse the imperfection sizes and scatters. The fatigue test results of the cruciform joints show, that undercuts lead to lower fatigue strength at the weld toe, as was to be expected. When reducing the fatigue resistance, undercuts could however be tolerated at secondary structural elements or marginally fatigue stressed components. A lack of penetration and/or root fusion can be tolerated, if the fatigue verification at the weld root is performed. A verification at the weld root can only be neglected in case of lack of penetration and/or root fusion, if the weld toe is decisive due to undercuts or large enough weld angels, here shown for 45°.

Graph convolutional networks in language and vision: A survey

Graph convolutional networks (GCNs) have a strong ability to learn graph representation and have achieved good performance in a range of applications, including social relationship analysis, biological information processing, natural language processing (NLP), computer vision (CV), and so on. In recent years, the application of GCNs in natural language processing and computer vision has attracted substantial interest from researchers, as a result of which many studies based on GCNs have emerged in the fields of natural language processing and computer vision. However, to the best of our knowledge, a comprehensive survey of GCN application in natural language processing and computer vision has not yet been conducted. Accordingly, this survey presents a comprehensive review of the principles of GCNs and its applications in these two fields. First, we summarize the principles of the two types of GCNs, namely spatial methods and spectral methods. Then we divide GCN applications into two categories: natural language processing and computer vision. Subsequently, we present multiple applications from each category in detail. Finally, we outline the limitations of GCNs and discuss possible future research directions.

Analysis of fatigue cracking of orthotropic steel decks using XFEM

The cutouts of diaphragms and the weld tips of the U-rib-to-diaphragm connections are the very complex structural details in orthotropic steel decks, and are accordingly highly vulnerable to fatigue cracking. The majority of fatigue cracks in these details appear as a result of the stress concentrations and initial weld defects. To determine the reason for the high probability of occurrence of fatigue cracks at these types of fatigue detail categories, a numerical model was established using the extended finite element method (XFEM) to analyze the cracking of diaphragm of an experimental 1:2 scaled deck segment model. This numerical analysis indicated that the cracks that appear at the diaphragm cutouts and near the weld toes of U-rib-to-diaphragm connections are induced by the relative horizontal displacement between the deck plate and diaphragm along the longitudinal axis of the bridge rather than by vertical deflection or moment due to live load. Additionally, stress concentrations play an important role in the initiation and propagation of cracks in these regions. The contribution of the out-of-plane displacement to fatigue cracking may be the most significant, and can frequently result in crack initiation and propagation prior to any cracking caused by vertical deflection or in-plane stresses. The XFEM approach is a useful numerical method for modelling the fatigue cracking of orthotropic steel decks.

Predicting CRISPR/Cas9 Repair Outcomes by Attention-Based Deep Learning Framework.

As a simple and programmable nuclease-based genome editing tool, the CRISPR/Cas9 system has been widely used in target-gene repair and gene-expression regulation. The DNA mutation generated by CRISPR/Cas9-mediated double-strand breaks determines its biological and phenotypic effects. Experiments have demonstrated that CRISPR/Cas9-generated cellular-repair outcomes depend on local sequence features. Therefore, the repair outcomes after DNA break can be predicted by sequences near the cleavage sites. However, existing prediction methods rely on manually constructed features or insufficiently detailed prediction labels. They cannot satisfy clinical-level-prediction accuracy, which limit the performance of these models to existing knowledge about CRISPR/Cas9 editing. We predict 557 repair labels of DNA, covering the vast majority of Cas9-generated mutational outcomes, and build a deep learning model called Apindel, to predict CRISPR/Cas9 editing outcomes. Apindel, automatically, trains the sequence features of DNA with the GloVe model, introduces location information through Positional Encoding (PE), and embeds the trained-word vector matrixes into a deep learning model, containing BiLSTM and the Attention mechanism. Apindel has better performance and more detailed prediction categories than the most advanced DNA-mutation-predicting models. It, also, reveals that nucleotides at different positions relative to the cleavage sites have different influences on CRISPR/Cas9 editing outcomes.

Joint effects of general population polygenic risk scores (PRS) and radiation treatment on subsequent neoplasm risk among childhood cancer survivors: A report from the Childhood Cancer Survivor Study (CCSS).

10008 Background: We examined whether PRS from general population studies are associated with risk of subsequent neoplasms (SNs) in childhood cancer survivors and evaluated joint associations between PRS and radiation treatment (RT), an established SN risk factor. Methods: Common genetic variants associated with risk of basal cell carcinoma (BCC), breast cancer, thyroid cancer, squamous cell carcinoma (SCC) or melanoma in general population studies were used to calculate cancer specific PRS among 5,911 5 year cancer survivors diagnosed < 21 years of age and between 1970-1986 in the CCSS. We examined associations between each PRS and SN risk using conditional logistic regression in nested case control studies, with incidence density sampling and matching on childhood cancer type, age at diagnosis, sex, RT dose to the SN site, and chemotherapy exposure. Further analyses matching only on non-treatment factors assessed joint associations considering potential combinations of PRS and RT exposure. We calculated the relative excess risk due to interaction (RERI) to determine whether joint associations were consistent with additivity of the individual risk factors (RERI > 0 indicates a more-than-additive joint association). Results: Among survivors (median age at follow-up 40 years, range 3-67; 62% exposed to RT), cancer specific PRS were associated with risk of subsequent BCC (N = 626; quartile 4 versus 1, OR [95% CI] = 1.9 [1.5-2.4]), breast cancer (N = 277; 4.5 [2.8-7.1]), thyroid cancer (N = 149; 1.9 [1.2-3.1]), and melanoma (N = 76; 2.7 [1.3-5.6]). Both PRS and RT were independently associated with SN risk, but joint analyses using a common reference group (PRS < median, RT < 1 Gy) found that both risk factors together resulted in more-than-additive increases in risk of BCC (RERI [95% CI] = 6.9 [2.0-11.8], breast cancer (6.6 [2.2-10.1], and thyroid cancer (4.8 [0.5-9.2]). Specifically, BCC risk was increased 28.8-fold for both PRS ≥ median and RT ≥ 1 Gy together, but only 3.3-fold for PRS ≥ median alone and 19.7-fold for RT ≥ 1 Gy alone. Similarly, breast cancer risk was increased 14.1-fold for both risk factors together, 2.5-fold for PRS ≥ median alone, and 6.5-fold for RT ≥ 1 Gy alone, and thyroid cancer risk was increased 12.3-fold for both risk factors together, 2.4-fold for PRS ≥ median alone, and 6.0-fold for RT ≥ 1 Gy alone. In joint analyses using more detailed RT categories, we found more-than-additive joint associations at both low and high RT doses. Conclusions: General population PRS were associated with SN risks after childhood cancer. More-than-additive increased risks with the combination of PRS and RT suggest that established markers of genetic susceptibility remain important in the context of treatment-related risks and may be useful in further refining risk assessment and follow-up guidelines for survivors.

Estimates of COVID‐19 vaccine uptake in major occupational groups and detailed occupational categories in the United States, April–May 2021

BackgroundWhile other studies have reported estimates of COVID‐19 vaccine uptake by broad occupational group, little is known about vaccine uptake by detailed occupational category.MethodsData on COVID‐19 vaccination were provided by US adults ages ≥18 years old who responded to the Facebook/Delphi Group COVID‐19 Trends and Impact Survey (Delphi US CTIS) in April–May 2021, reported working for pay in the past 4 weeks, and answered questions about their COVID‐19 vaccine status. Percentages of occupational groups reporting having had at least one COVID‐19 vaccination were weighted to resemble the US general population and calculated for 23 major occupational groups and 120 detailed occupational categories in 15 major groups.ResultsCOVID‐19 vaccine uptake for all 828,401 working adult respondents was 73.6%. Uptake varied considerably across the 23 major occupational groups, from 45.7% for Construction and Extraction to 87.9% for Education, Training, and Library. Percentage vaccinated was also very low for Installation, Maintenance, and Repair at 52.1% and Farming, Fishing, and Forestry at 53.9%. Among the 120 detailed occupational categories, the highest percentage vaccinated was 93.9% for Postsecondary Teacher and the three lowest values were 39.1% for Any Extraction Worker in Oil, Gas, Mining, or Quarrying; 40.1% for Vehicle or Mobile Equipment Mechanic, Installer, or Repairer; and 42.0% for Any Construction Trades Worker.ConclusionLow vaccination percentages were seen in many US occupations by the end of May 2021, early in the period of widespread availability of vaccines for adults. These findings could help inform the deployment of occupation‐specific vaccinepromotion activities during future viral epidemics and pandemics.