Maysam Orouskhani, Sarwesh Rauniyar, Norma Morella, Daniel Lachance, Samuel S Minot, Neelendu Dey
{"title":"深度学习成像分析识别与致癌物产生相关的细菌代谢状态。","authors":"Maysam Orouskhani, Sarwesh Rauniyar, Norma Morella, Daniel Lachance, Samuel S Minot, Neelendu Dey","doi":"10.1007/s44352-025-00006-1","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Colorectal cancer (CRC) is a globally prevalent cancer. Emerging research implicates the gut microbiome in CRC pathogenesis. Bacteria such as <i>Clostridium scindens</i> can produce the carcinogenic bile acid deoxycholic acid (DCA). It is unknown whether imaging methods can differentiate DCA-producing and DCA-non-producing <i>C. scindens</i> cells.</p><p><strong>Methods: </strong>Light microscopy images of anaerobically cultured <i>C. scindens</i> in four conditions were acquired at 100× magnification using the Tissue FAX system: <i>C. scindens</i> in media alone (DCA-non-producing state), <i>C. scindens</i> in media with cholic acid (DCA-producing state), or <i>C. scindens</i> in co-culture with one of two <i>Bacteroides</i> species (intermediate DCA production states). We evaluated three approaches: whole-image classification, per-cell classification, and image segmentation-based classification. For whole-image classification, we used a custom Convolutional Neural Network (CNN), pre-trained DenseNet, pre-trained ResNet, and ResNet enhanced by integrating the Digital Images of Bacterial Species (DIBaS) dataset. For cell detection and classification, we applied thresholding (OTSU or adaptive thresholding) followed by a ResNet model. Finally, image segmentation-based classification was performed using nnU-Net.</p><p><strong>Results: </strong>For whole-image analysis, DIBaS-enhanced ResNet models achieved the best performance in distinguishing <i>C. scindens</i> states in monoculture (accuracy 0.89 ± 0.006) and in co-cultures (accuracy 0.86 ± 0.004). Per-cell analysis was optimal at a C constant value of 3, with the ResNet model achieving 62-74% accuracy for <i>C. scindens</i> states in monoculture. Segmentation-based analysis using nnU-Net resulted in Dice coefficients of 87% for <i>C. scindens</i> and 74-76% for the <i>Bacteroides</i> species.</p><p><strong>Conclusions: </strong>This study demonstrates feasibility of image-based deep learning models in identifying health-relevant gut bacterial metabolic states.</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s44352-025-00006-1.</p>","PeriodicalId":520461,"journal":{"name":"Discover imaging","volume":"2 1","pages":"2"},"PeriodicalIF":0.0000,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11912549/pdf/","citationCount":"0","resultStr":"{\"title\":\"Deep learning imaging analysis to identify bacterial metabolic states associated with carcinogen production.\",\"authors\":\"Maysam Orouskhani, Sarwesh Rauniyar, Norma Morella, Daniel Lachance, Samuel S Minot, Neelendu Dey\",\"doi\":\"10.1007/s44352-025-00006-1\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Colorectal cancer (CRC) is a globally prevalent cancer. Emerging research implicates the gut microbiome in CRC pathogenesis. Bacteria such as <i>Clostridium scindens</i> can produce the carcinogenic bile acid deoxycholic acid (DCA). It is unknown whether imaging methods can differentiate DCA-producing and DCA-non-producing <i>C. scindens</i> cells.</p><p><strong>Methods: </strong>Light microscopy images of anaerobically cultured <i>C. scindens</i> in four conditions were acquired at 100× magnification using the Tissue FAX system: <i>C. scindens</i> in media alone (DCA-non-producing state), <i>C. scindens</i> in media with cholic acid (DCA-producing state), or <i>C. scindens</i> in co-culture with one of two <i>Bacteroides</i> species (intermediate DCA production states). We evaluated three approaches: whole-image classification, per-cell classification, and image segmentation-based classification. For whole-image classification, we used a custom Convolutional Neural Network (CNN), pre-trained DenseNet, pre-trained ResNet, and ResNet enhanced by integrating the Digital Images of Bacterial Species (DIBaS) dataset. For cell detection and classification, we applied thresholding (OTSU or adaptive thresholding) followed by a ResNet model. Finally, image segmentation-based classification was performed using nnU-Net.</p><p><strong>Results: </strong>For whole-image analysis, DIBaS-enhanced ResNet models achieved the best performance in distinguishing <i>C. scindens</i> states in monoculture (accuracy 0.89 ± 0.006) and in co-cultures (accuracy 0.86 ± 0.004). Per-cell analysis was optimal at a C constant value of 3, with the ResNet model achieving 62-74% accuracy for <i>C. scindens</i> states in monoculture. Segmentation-based analysis using nnU-Net resulted in Dice coefficients of 87% for <i>C. scindens</i> and 74-76% for the <i>Bacteroides</i> species.</p><p><strong>Conclusions: </strong>This study demonstrates feasibility of image-based deep learning models in identifying health-relevant gut bacterial metabolic states.</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s44352-025-00006-1.</p>\",\"PeriodicalId\":520461,\"journal\":{\"name\":\"Discover imaging\",\"volume\":\"2 1\",\"pages\":\"2\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2025-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11912549/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Discover imaging\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1007/s44352-025-00006-1\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/3/10 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Discover imaging","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/s44352-025-00006-1","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/3/10 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}
Deep learning imaging analysis to identify bacterial metabolic states associated with carcinogen production.
Background: Colorectal cancer (CRC) is a globally prevalent cancer. Emerging research implicates the gut microbiome in CRC pathogenesis. Bacteria such as Clostridium scindens can produce the carcinogenic bile acid deoxycholic acid (DCA). It is unknown whether imaging methods can differentiate DCA-producing and DCA-non-producing C. scindens cells.
Methods: Light microscopy images of anaerobically cultured C. scindens in four conditions were acquired at 100× magnification using the Tissue FAX system: C. scindens in media alone (DCA-non-producing state), C. scindens in media with cholic acid (DCA-producing state), or C. scindens in co-culture with one of two Bacteroides species (intermediate DCA production states). We evaluated three approaches: whole-image classification, per-cell classification, and image segmentation-based classification. For whole-image classification, we used a custom Convolutional Neural Network (CNN), pre-trained DenseNet, pre-trained ResNet, and ResNet enhanced by integrating the Digital Images of Bacterial Species (DIBaS) dataset. For cell detection and classification, we applied thresholding (OTSU or adaptive thresholding) followed by a ResNet model. Finally, image segmentation-based classification was performed using nnU-Net.
Results: For whole-image analysis, DIBaS-enhanced ResNet models achieved the best performance in distinguishing C. scindens states in monoculture (accuracy 0.89 ± 0.006) and in co-cultures (accuracy 0.86 ± 0.004). Per-cell analysis was optimal at a C constant value of 3, with the ResNet model achieving 62-74% accuracy for C. scindens states in monoculture. Segmentation-based analysis using nnU-Net resulted in Dice coefficients of 87% for C. scindens and 74-76% for the Bacteroides species.
Conclusions: This study demonstrates feasibility of image-based deep learning models in identifying health-relevant gut bacterial metabolic states.
Supplementary information: The online version contains supplementary material available at 10.1007/s44352-025-00006-1.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
