Pub Date : 2025-11-01DOI: 10.1088/1742-6596/3140/9/092013
G Petrou, Y Wang, Z Chalabi, S C Hsu, E Hutchinson, J Milner, P Symonds, M Davies
This paper explores the impact of different approaches to aggregating the indoor concentration of fine particulate matter (PM2.5) in a case study home. Indoor- and outdoor-sourced PM2.5 was modelled in CONTAM-EnergyPlus for a bungalow occupied by a family of four in Plymouth, England. Simulations were conducted assuming energy efficiency levels typical of a 1950s home and following retrofit. Pollutants were modelled at a 5-min temporal resolution in bedrooms, kitchen and living room and aggregated according to four metrics: (i) household arithmetic mean concentration, (ii) household time-weighted mean concentration, (iii) arithmetic mean of individual exposure, and (iv) household arithmetic mean exposure. Comparing the household metrics revealed differences of up to 50.0 % (3.6 μg/m3) for the pre-retrofit model, which remained largely consistent following retrofit. When comparing against individual exposure, differences were observed for all three metrics and reached 55.6% (9.1 μg/m3) for the pre-retrofit model, and 50.6% (8.7 μg/m3) for the post-retrofit model. Approaches (i) and (ii) consistently underpredicted individual exposure. Further, the differences were greatest for the time-weighted mean method, suggesting that taking into consideration the total time that each room is occupied but not when it is occupied will not necessarily provide a more accurate description of occupant exposure compared to the simple arithmetic mean. To better represent individual exposure, data on occupant presence is required.
{"title":"Comparing methods for aggregating indoor air pollutant concentration over space and time.","authors":"G Petrou, Y Wang, Z Chalabi, S C Hsu, E Hutchinson, J Milner, P Symonds, M Davies","doi":"10.1088/1742-6596/3140/9/092013","DOIUrl":"10.1088/1742-6596/3140/9/092013","url":null,"abstract":"<p><p>This paper explores the impact of different approaches to aggregating the indoor concentration of fine particulate matter (PM<sub>2.5</sub>) in a case study home. Indoor- and outdoor-sourced PM<sub>2.5</sub> was modelled in CONTAM-EnergyPlus for a bungalow occupied by a family of four in Plymouth, England. Simulations were conducted assuming energy efficiency levels typical of a 1950s home and following retrofit. Pollutants were modelled at a 5-min temporal resolution in bedrooms, kitchen and living room and aggregated according to four metrics: (i) household arithmetic mean concentration, (ii) household time-weighted mean concentration, (iii) arithmetic mean of individual exposure, and (iv) household arithmetic mean exposure. Comparing the household metrics revealed differences of up to 50.0 % (3.6 μg/m<sup>3</sup>) for the pre-retrofit model, which remained largely consistent following retrofit. When comparing against individual exposure, differences were observed for all three metrics and reached 55.6% (9.1 μg/m<sup>3</sup>) for the pre-retrofit model, and 50.6% (8.7 μg/m<sup>3</sup>) for the post-retrofit model. Approaches (i) and (ii) consistently underpredicted individual exposure. Further, the differences were greatest for the time-weighted mean method, suggesting that taking into consideration the total time that each room is occupied but not when it is occupied will not necessarily provide a more accurate description of occupant exposure compared to the simple arithmetic mean. To better represent individual exposure, data on occupant presence is required.</p>","PeriodicalId":520472,"journal":{"name":"Journal of physics. Conference series","volume":"3140 9","pages":"092013-92013"},"PeriodicalIF":0.0,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7618645/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-01DOI: 10.1088/1742-6596/2722/1/012012
Can Cui, Ruining Deng, Quan Liu, Tianyuan Yao, Shunxing Bao, Lucas W Remedios, Bennett A Landman, Yucheng Tang, Yuankai Huo
The Segment Anything Model (SAM) is a recently proposed prompt-based segmentation model in a generic zero-shot segmentation approach. With the zero-shot segmentation capacity, SAM achieved impressive flexibility and precision on various segmentation tasks. However, the current pipeline requires manual prompts during the inference stage, which is still resource intensive for biomedical image segmentation. In this paper, instead of using prompts during the inference stage, we introduce a pipeline that utilizes the SAM, called all-in-SAM, through the entire AI development workflow (from annotation generation to model finetuning) without requiring manual prompts during the inference stage. Specifically, SAM is first employed to generate pixel-level annotations from weak prompts (e.g., points, bounding box). Then, the pixel-level annotations are used to finetune the SAM segmentation model rather than training from scratch. Our experimental results reveal two key findings: 1) the proposed pipeline surpasses the state-of-the-art methods in a nuclei segmentation task on the public Monuseg dataset, and 2) the utilization of weak and few annotations for SAM finetuning achieves competitive performance compared to using strong pixelwise annotated data.
{"title":"All-in-SAM: from Weak Annotation to Pixel-wise Nuclei Segmentation with Prompt-based Finetuning.","authors":"Can Cui, Ruining Deng, Quan Liu, Tianyuan Yao, Shunxing Bao, Lucas W Remedios, Bennett A Landman, Yucheng Tang, Yuankai Huo","doi":"10.1088/1742-6596/2722/1/012012","DOIUrl":"10.1088/1742-6596/2722/1/012012","url":null,"abstract":"<p><p>The Segment Anything Model (SAM) is a recently proposed prompt-based segmentation model in a generic zero-shot segmentation approach. With the zero-shot segmentation capacity, SAM achieved impressive flexibility and precision on various segmentation tasks. However, the current pipeline requires manual prompts during the inference stage, which is still resource intensive for biomedical image segmentation. In this paper, instead of using prompts during the inference stage, we introduce a pipeline that utilizes the SAM, called all-in-SAM, through the entire AI development workflow (from annotation generation to model finetuning) without requiring manual prompts during the inference stage. Specifically, SAM is first employed to generate pixel-level annotations from weak prompts (e.g., points, bounding box). Then, the pixel-level annotations are used to finetune the SAM segmentation model rather than training from scratch. Our experimental results reveal two key findings: 1) the proposed pipeline surpasses the state-of-the-art methods in a nuclei segmentation task on the public Monuseg dataset, and 2) the utilization of weak and few annotations for SAM finetuning achieves competitive performance compared to using strong pixelwise annotated data.</p>","PeriodicalId":520472,"journal":{"name":"Journal of physics. Conference series","volume":"2722 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11925546/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143672237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号