Attention-effective multiple instance learning on weakly stem cell colony segmentation
The detection of induced pluripotent stem cell (iPSC) colonies often needs the precise extraction of the colony features. However, existing computerized systems relied on segmentation of contours by preprocessing for classifying the colony conditions were task-extensive. To maximize the efficiency i...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-02-01
|
| Series: | Intelligent Systems with Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667305323000121 |
| _version_ | 1811170983133839360 |
|---|---|
| author | Novanto Yudistira Muthu Subash Kavitha Jeny Rajan Takio Kurita |
| author_facet | Novanto Yudistira Muthu Subash Kavitha Jeny Rajan Takio Kurita |
| author_sort | Novanto Yudistira |
| collection | DOAJ |
| description | The detection of induced pluripotent stem cell (iPSC) colonies often needs the precise extraction of the colony features. However, existing computerized systems relied on segmentation of contours by preprocessing for classifying the colony conditions were task-extensive. To maximize the efficiency in categorizing colony conditions, we propose a multiple instance learning (MIL) in weakly supervised settings. It is designed in a single model to produce weak segmentation and classification of colonies without using finely labeled samples. As a single model, we employ a U-net-like convolution neural network (CNN) to train on binary image-level labels for MIL colonies classification. Furthermore, to specify the object of interest we used a simple post-processing method. The proposed approach is compared over conventional methods using five-fold cross-validation and receiver operating characteristic (ROC) curve. The maximum accuracy of the MIL-net is 95%, which is 15% higher than the conventional methods. Furthermore, the ability to interpret the location of the iPSC colonies based on the image level label without using a pixel-wise ground truth image is more appealing and cost-effective in colony condition recognition. |
| first_indexed | 2024-04-10T17:05:52Z |
| format | Article |
| id | doaj.art-1a73c9f0b0fb4a99ae7edac403de6d05 |
| institution | Directory Open Access Journal |
| issn | 2667-3053 |
| language | English |
| last_indexed | 2024-04-10T17:05:52Z |
| publishDate | 2023-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Intelligent Systems with Applications |
| spelling | doaj.art-1a73c9f0b0fb4a99ae7edac403de6d052023-02-06T04:06:32ZengElsevierIntelligent Systems with Applications2667-30532023-02-0117200187Attention-effective multiple instance learning on weakly stem cell colony segmentationNovanto Yudistira0Muthu Subash Kavitha1Jeny Rajan2Takio Kurita3Informatics Department, Faculty of Computer Science, Brawijaya University, Jalan Veteran 8, Malang, 65145, Malang, IndonesiaSchool of Information and Data Sciences, Nagasaki University, Nagasaki city 852-8521, Nagasaki, Japan; Corresponding author.Department of Computer Science and Engineering, National Institute of Technology, Karnataka, Surathkal, IndiaGraduate School of Advanced Science and Engineering, Hiroshima University, Hiroshima, JapanThe detection of induced pluripotent stem cell (iPSC) colonies often needs the precise extraction of the colony features. However, existing computerized systems relied on segmentation of contours by preprocessing for classifying the colony conditions were task-extensive. To maximize the efficiency in categorizing colony conditions, we propose a multiple instance learning (MIL) in weakly supervised settings. It is designed in a single model to produce weak segmentation and classification of colonies without using finely labeled samples. As a single model, we employ a U-net-like convolution neural network (CNN) to train on binary image-level labels for MIL colonies classification. Furthermore, to specify the object of interest we used a simple post-processing method. The proposed approach is compared over conventional methods using five-fold cross-validation and receiver operating characteristic (ROC) curve. The maximum accuracy of the MIL-net is 95%, which is 15% higher than the conventional methods. Furthermore, the ability to interpret the location of the iPSC colonies based on the image level label without using a pixel-wise ground truth image is more appealing and cost-effective in colony condition recognition.http://www.sciencedirect.com/science/article/pii/S2667305323000121Multiple instanceWeakly supervised segmentationColonyAnnotationInference |
| spellingShingle | Novanto Yudistira Muthu Subash Kavitha Jeny Rajan Takio Kurita Attention-effective multiple instance learning on weakly stem cell colony segmentation Intelligent Systems with Applications Multiple instance Weakly supervised segmentation Colony Annotation Inference |
| title | Attention-effective multiple instance learning on weakly stem cell colony segmentation |
| title_full | Attention-effective multiple instance learning on weakly stem cell colony segmentation |
| title_fullStr | Attention-effective multiple instance learning on weakly stem cell colony segmentation |
| title_full_unstemmed | Attention-effective multiple instance learning on weakly stem cell colony segmentation |
| title_short | Attention-effective multiple instance learning on weakly stem cell colony segmentation |
| title_sort | attention effective multiple instance learning on weakly stem cell colony segmentation |
| topic | Multiple instance Weakly supervised segmentation Colony Annotation Inference |
| url | http://www.sciencedirect.com/science/article/pii/S2667305323000121 |
| work_keys_str_mv | AT novantoyudistira attentioneffectivemultipleinstancelearningonweaklystemcellcolonysegmentation AT muthusubashkavitha attentioneffectivemultipleinstancelearningonweaklystemcellcolonysegmentation AT jenyrajan attentioneffectivemultipleinstancelearningonweaklystemcellcolonysegmentation AT takiokurita attentioneffectivemultipleinstancelearningonweaklystemcellcolonysegmentation |