Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data
Outdoor ultrafine particles (UFPs) (<0.1 µm) may have an important impact on public health but exposure assessment remains a challenge in epidemiological studies. We developed a novel method of estimating spatiotemporal variations in outdoor UFP number concentrations and particle diameters using...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-11-01
|
| Series: | Environment International |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0160412020319991 |
| _version_ | 1828822385604689920 |
|---|---|
| author | Kris Y. Hong Pedro O. Pinheiro Scott Weichenthal |
| author_facet | Kris Y. Hong Pedro O. Pinheiro Scott Weichenthal |
| author_sort | Kris Y. Hong |
| collection | DOAJ |
| description | Outdoor ultrafine particles (UFPs) (<0.1 µm) may have an important impact on public health but exposure assessment remains a challenge in epidemiological studies. We developed a novel method of estimating spatiotemporal variations in outdoor UFP number concentrations and particle diameters using street-level images and audio data in Montreal, Canada. As a secondary aim, we also developed models for noise. Convolutional neural networks were first trained to predict 10-second average UFP/noise parameters using a large database of images and audio spectrogram data paired with measurements collected between April 2019 and February 2020. Final multivariable linear regression and generalized additive models were developed to predict 5-minute average UFP/noise parameters including covariates from deep learning models based on image and audio data along with outdoor temperature and wind speed. The best performing final models had mean cross-validation R2 values of 0.677 and 0.523 for UFP number concentrations and 0.825 and 0.735 for UFP size using two different test sets. Audio predictions from deep learning models were stronger predictors of spatiotemporal variations in UFP parameters than predictions based on street-level images; this was not explained only by noise levels captured in the audio signal. All final noise models had R2 values above 0.90. Collectively, our findings suggest that street-level images and audio data can be used to estimate spatiotemporal variations in outdoor UFPs and noise. This approach may be useful in developing exposure models over broad spatial scales and such models can be regularly updated to expand generalizability as more measurements become available. |
| first_indexed | 2024-12-12T13:12:52Z |
| format | Article |
| id | doaj.art-1649970a3a4f434686b1f0e6a8f7b88d |
| institution | Directory Open Access Journal |
| issn | 0160-4120 |
| language | English |
| last_indexed | 2024-12-12T13:12:52Z |
| publishDate | 2020-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Environment International |
| spelling | doaj.art-1649970a3a4f434686b1f0e6a8f7b88d2022-12-22T00:23:28ZengElsevierEnvironment International0160-41202020-11-01144106044Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio dataKris Y. Hong0Pedro O. Pinheiro1Scott Weichenthal2McGill University, Department of Epidemiology, Biostatistics and Occupational Health, Montreal, QC H3A 1A3, Canada; Element AI, 6650 Saint Urbain, Suite #500, Montreal, QC H2S 3G9, CanadaElement AI, 6650 Saint Urbain, Suite #500, Montreal, QC H2S 3G9, CanadaMcGill University, Department of Epidemiology, Biostatistics and Occupational Health, Montreal, QC H3A 1A3, Canada; Corresponding author at: Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, 1110 avenue des Pins Ouest, Montreal, QC H3A 1A3, Canada.Outdoor ultrafine particles (UFPs) (<0.1 µm) may have an important impact on public health but exposure assessment remains a challenge in epidemiological studies. We developed a novel method of estimating spatiotemporal variations in outdoor UFP number concentrations and particle diameters using street-level images and audio data in Montreal, Canada. As a secondary aim, we also developed models for noise. Convolutional neural networks were first trained to predict 10-second average UFP/noise parameters using a large database of images and audio spectrogram data paired with measurements collected between April 2019 and February 2020. Final multivariable linear regression and generalized additive models were developed to predict 5-minute average UFP/noise parameters including covariates from deep learning models based on image and audio data along with outdoor temperature and wind speed. The best performing final models had mean cross-validation R2 values of 0.677 and 0.523 for UFP number concentrations and 0.825 and 0.735 for UFP size using two different test sets. Audio predictions from deep learning models were stronger predictors of spatiotemporal variations in UFP parameters than predictions based on street-level images; this was not explained only by noise levels captured in the audio signal. All final noise models had R2 values above 0.90. Collectively, our findings suggest that street-level images and audio data can be used to estimate spatiotemporal variations in outdoor UFPs and noise. This approach may be useful in developing exposure models over broad spatial scales and such models can be regularly updated to expand generalizability as more measurements become available.http://www.sciencedirect.com/science/article/pii/S0160412020319991Ultrafine particlesNoiseDeep learningImagesAudio |
| spellingShingle | Kris Y. Hong Pedro O. Pinheiro Scott Weichenthal Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data Environment International Ultrafine particles Noise Deep learning Images Audio |
| title | Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data |
| title_full | Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data |
| title_fullStr | Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data |
| title_full_unstemmed | Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data |
| title_short | Predicting outdoor ultrafine particle number concentrations, particle size, and noise using street-level images and audio data |
| title_sort | predicting outdoor ultrafine particle number concentrations particle size and noise using street level images and audio data |
| topic | Ultrafine particles Noise Deep learning Images Audio |
| url | http://www.sciencedirect.com/science/article/pii/S0160412020319991 |
| work_keys_str_mv | AT krisyhong predictingoutdoorultrafineparticlenumberconcentrationsparticlesizeandnoiseusingstreetlevelimagesandaudiodata AT pedroopinheiro predictingoutdoorultrafineparticlenumberconcentrationsparticlesizeandnoiseusingstreetlevelimagesandaudiodata AT scottweichenthal predictingoutdoorultrafineparticlenumberconcentrationsparticlesizeandnoiseusingstreetlevelimagesandaudiodata |