{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T03:41:36Z","timestamp":1772163696095,"version":"3.50.1"},"reference-count":17,"publisher":"BMJ","issue":"9","license":[{"start":{"date-parts":[[2022,9,19]],"date-time":"2022-09-19T00:00:00Z","timestamp":1663545600000},"content-version":"unspecified","delay-in-days":18,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100010269","name":"Wellcome Trust","doi-asserted-by":"publisher","award":["214185\/Z\/18\/Z"],"award-info":[{"award-number":["214185\/Z\/18\/Z"]}],"id":[{"id":"10.13039\/100010269","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Open"],"published-print":{"date-parts":[[2022,9]]},"abstract":"<jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>During the COVID-19 pandemic, convolutional neural networks (CNNs) have been used in clinical medicine (eg, X-rays classification). Whether CNNs could inform the epidemiology of COVID-19 classifying street images according to COVID-19 risk is unknown, yet it could pinpoint high-risk places and relevant features of the built environment. In a feasibility study, we trained CNNs to classify the area surrounding bus stops (Lima, Peru) into moderate or extreme COVID-19 risk.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Design<\/jats:title>\n                    <jats:p>CNN analysis based on images from bus stops and the surrounding area. We used transfer learning and updated the output layer of five CNNs: NASNetLarge, InceptionResNetV2, Xception, ResNet152V2 and ResNet101V2. We chose the best performing CNN, which was further tuned. We used GradCam to understand the classification process.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Setting<\/jats:title>\n                    <jats:p>Bus stops from Lima, Peru. We used five images per bus stop.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Primary and secondary outcome measures<\/jats:title>\n                    <jats:p>Bus stop images were classified according to COVID-19 risk into two labels: moderate or extreme.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>NASNetLarge outperformed the other CNNs except in the recall metric for the moderate label and in the precision metric for the extreme label; the ResNet152V2 performed better in these two metrics (85% vs 76% and 63% vs 60%, respectively). The NASNetLarge was further tuned. The best recall (75%) and F1 score (65%) for the extreme label were reached with data augmentation techniques. Areas close to buildings or with people were often classified as extreme risk.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>This feasibility study showed that CNNs have the potential to classify street images according to levels of COVID-19 risk. In addition to applications in clinical medicine, CNNs and street images could advance the epidemiology of COVID-19 at the population level.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjopen-2022-063411","type":"journal-article","created":{"date-parts":[[2022,9,19]],"date-time":"2022-09-19T13:10:29Z","timestamp":1663593029000},"page":"e063411","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":1,"title":["Street images classification according to COVID-19 risk in Lima, Peru: a convolutional neural networks feasibility analysis"],"prefix":"10.1136","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2090-1856","authenticated-orcid":false,"given":"Rodrigo M","family":"Carrillo-Larco","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2990-7090","authenticated-orcid":false,"given":"Manuel","family":"Castillo-Cara","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jose Francisco","family":"Hern\u00e1ndez Santa Cruz","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"239","published-online":{"date-parts":[[2022,9,19]]},"reference":[{"key":"2022091910053272000_12.9.e063411.1","doi-asserted-by":"publisher","DOI":"10.1155\/2021\/6677314"},{"key":"2022091910053272000_12.9.e063411.2","doi-asserted-by":"crossref","DOI":"10.3389\/fcvm.2021.638011","article-title":"Application of machine learning in diagnosis of COVID-19 through X-ray and CT images: a scoping review","volume":"8","author":"Mohammad-Rahimi","year":"2021","journal-title":"Front Cardiovasc Med"},{"key":"2022091910053272000_12.9.e063411.3","doi-asserted-by":"crossref","unstructured":"Montazeri M , ZahediNasab R , Farahani A , et al . Machine learning models for image-based diagnosis and prognosis of COVID-19: systematic review 2021;9:e25181.","DOI":"10.2196\/25181"},{"key":"2022091910053272000_12.9.e063411.4","doi-asserted-by":"crossref","unstructured":"Nguyen QC , Huang Y , Kumar A , et al . Using 164 million Google street view images to derive built environment predictors of COVID-19 cases. Int J Environ Res Public Health 2020;17:6359.doi:10.3390\/ijerph17176359","DOI":"10.3390\/ijerph17176359"},{"key":"2022091910053272000_12.9.e063411.5","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-019-42036-w","article-title":"Measuring social, environmental and health inequalities using deep learning and street imagery","volume":"9","author":"Suel","year":"2019","journal-title":"Sci Rep"},{"key":"2022091910053272000_12.9.e063411.6","doi-asserted-by":"crossref","first-page":"3429","DOI":"10.3390\/rs14143429","article-title":"What you see is what you breathe? estimating air pollution spatial variation using Street-Level imagery","volume":"14","author":"Suel","year":"2022","journal-title":"Remote Sens"},{"key":"2022091910053272000_12.9.e063411.7","doi-asserted-by":"crossref","DOI":"10.1016\/j.healthplace.2020.102428","article-title":"Objective scoring of streetscape walkability related to leisure walking: statistical modeling approach with semantic segmentation of Google street view images","volume":"66","author":"Nagata","year":"2020","journal-title":"Health Place"},{"key":"2022091910053272000_12.9.e063411.8","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1177\/0033354920968799","article-title":"Leveraging 31 million Google street view images to characterize built environments and examine County health outcomes","volume":"136","author":"Nguyen","year":"2021","journal-title":"Public Health Rep"},{"key":"2022091910053272000_12.9.e063411.9","doi-asserted-by":"publisher","DOI":"10.1136\/jech-2017-209456"},{"key":"2022091910053272000_12.9.e063411.10","unstructured":"Autoridad de Transporte Urbano para Lima y Callao (ATU) . Paraderos con Riesgo de COVID - 19, 2022. Available: https:\/\/sistemas.atu.gob.pe\/paraderosCOVID"},{"key":"2022091910053272000_12.9.e063411.11","unstructured":"jmcastagnetto . Data from: lima-atu-covid19-paraderos, 2022. Available: https:\/\/github.com\/jmcastagnetto\/lima-atu-covid19-paraderos"},{"key":"2022091910053272000_12.9.e063411.12","unstructured":"Keras applications, 2022. Available: https:\/\/keras.io\/api\/applications\/"},{"key":"2022091910053272000_12.9.e063411.13","doi-asserted-by":"crossref","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","article-title":"Grad-CAM: visual explanations from deep networks via Gradient-Based localization","volume":"128","author":"Selvaraju","year":"2020","journal-title":"International Journal of Computer Vision"},{"key":"2022091910053272000_12.9.e063411.14","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2923-3"},{"key":"2022091910053272000_12.9.e063411.15","doi-asserted-by":"crossref","unstructured":"Ganaie M , Hu M . Ensemble deep learning: a review. arXiv preprint arXiv:210402395 2021.","DOI":"10.1016\/j.engappai.2022.105151"},{"key":"2022091910053272000_12.9.e063411.16","unstructured":"Hou L , C-P Y , Samaras D . Squared earth mover\u2019s distance-based loss for training deep neural networks. arXiv preprint arXiv:161105916 2016."},{"key":"2022091910053272000_12.9.e063411.17","unstructured":"Iandola FN , Han S , Moskewicz MW , et al . SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and. arXiv preprint arXiv:160207360 2016."}],"container-title":["BMJ Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjopen-2022-063411","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,19]],"date-time":"2022-09-19T13:10:45Z","timestamp":1663593045000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjopen.bmj.com\/lookup\/doi\/10.1136\/bmjopen-2022-063411"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,9]]},"references-count":17,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2022,9,19]]},"published-print":{"date-parts":[[2022,9]]}},"alternative-id":["10.1136\/bmjopen-2022-063411"],"URL":"https:\/\/doi.org\/10.1136\/bmjopen-2022-063411","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2021.09.06.21263188","asserted-by":"object"}]},"ISSN":["2044-6055","2044-6055"],"issn-type":[{"value":"2044-6055","type":"print"},{"value":"2044-6055","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,9]]}}}