Deep Learning based Segmentation Pipeline for Label-Free Phase-Contrast Microscopy Images

Deep Learning based Segmentation Pipeline
for Label-Free Phase-Contrast Microscopy Images
Aydin Ayanzadeh, Özden Yal¸cın Özuysal, Devrim Pesen Okvur, Sevgi
Önal, Beh¸cet U˘gur Töreyin, Devrim Ünay
Istanbul Technical University, Izmir Institute of Technology,
Izmir Demokrasi University
September 23, 2020
Aydin Ayanzadeh (SpC4ing Group./˙ITÜ) Cell Segmentation of PCM. September 23, 2020 1 / 20

Outline
Problem Motivation
Dataset
Dataset Preparation
Proposed Method
Evaluation Method
Conclusion
Future Work
References

Problem Motivation
Segmentation of medical images especially on microscopy images is
challenging.
segmentation make ease the process of cell behaviours.
Reduce the workload and provide oppurtunity for batch analysis of the
cells.

Literature Review
U-Net[1] and its variants
Maskrcnn[2]
Figure: Representation of U-Net Architecture[1].

Dataset
MDA MB-213
Invasive breast cancer cells that which has mesenchymal morphology.
Dataset collected and annotated with the help of experts.

Dataset Preparation
Figure: Workﬂow of Dataset Preparation.

Model
Figure: Representation of ResNet-18 Model[3].

Model
ResNet18-FPN
Applying Pretrained ResNet18 in the encoder of FPN.
Representation of ResNet-18-FPN Model.

Model
ResNet18-UNet
ResNet18 is replaced in the encoder of UNet.
Residual block is replaced in the decoder section.
Figure: Representation of ResNet-18-U-Net Model.

Training Methodology
The encoder has pre-train of ImageNet.
Test Time Augmentation(TTA)
Loss functions: Binary Cross Entropy loss + Dice Loss

Evaluation Metrics
Evaluation Metrics
Precision, Recall, Dice Coeﬃcient, Jaccard Index (IoU)
Precision =
ntp
nfp + ntp
(1)
Recall =
ntp
nfn + ntp
(2)
Jaccard(X, Y ) =
|X Y |
|X Y |
=
|X Y |
|X| + |Y | − |X Y |
(3)
Dice(X, Y ) =
2|X · Y |
|X| + |Y |
(4)

Experimental Results on MDA-MB-213
Table: Quantitative results of the segmentation methods.
Methods IoU Dice Precision Recall
EGT[4] 0.4225 0.5940 0.4574 0.8470
PHANTAST[5] 0.5465 0.7068 0.6823 0.7330
LinkNet [6] 0.8593 0.9235 0.9454 0.9025
U-Net [1] 0.8646 0.9274 0.9335 0.9214
Tip-Net[7] 0.8542 0.9214 0.9382 0.9052
ResNet18-UNet 0.8791 0.9357 0.9508 0.9211
ResNet18-FPN 0.8710 0.9311 0.9419 0.9206

Qualitative Results

Conclusion
Applying alternative encoder(ResNet18) on U-Net and FPN.
The model with alternative encoder and decoder is more robust to
outlier and boundary regions.
Applying pre-trained of ImageNet is successful in convergence to
better results.
Reduce the disparity in the encoder feature and the features that
propagate in the decoder of the U-Net architecture.

Future Work
Deploy methods for instance segmentation tasks and make it robust
for tracking purpose.
Applying the alternative encoder to gain better results.
Evaluate the performance methods on diﬀerent modality in medical
image datasets.
Extraction of the time-series dataset by the construction of lineage
relationships among the cells.

Any Question?
Thank You!

References I
Olaf Ronneberger, Philipp Fischer, and Thomas Brox.
U-net: Convolutional networks for biomedical image segmentation.
In Nassir Navab, Joachim Hornegger, William M. Wells, and
Alejandro F. Frangi, editors, Medical Image Computing and
Computer-Assisted Intervention – MICCAI 2015, pages 234–241,
Cham, 2015. Springer International Publishing.
Hsieh-Fu Tsai, Joanna Gajda, Tyler F.W. Sloan, Andrei Rares, and
Amy Q. Shen.
Usiigaci: Instance-aware cell tracking in stain-free phase contrast
microscopy enabled by machine learning.
SoftwareX, 9:230 – 237, 2019.
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun.
Deep residual learning for image recognition.
In Proceedings of the IEEE conference on computer vision and pattern
recognition, pages 770–778, 2016.

References II
Joe Chalfoun, M Majurski, A Peskin, Catherine Breen, Peter Bajcsy,
and M Brady.
Empirical gradient threshold technique for automated segmentation
across image modalities and cell lines.
Journal of microscopy, 260(1):86–99, 2015.
Nicolas Jaccard, Lewis D Griﬃn, Ana Keser, Rhys J Macown,
Alexandre Super, Farlan S Veraitch, and Nicolas Szita.
Automated method for the rapid and precise estimation of adherent
cell culture characteristics from phase contrast microscopy images.
Biotechnology and bioengineering, 111(3):504–517, 2014.
Abhishek Chaurasia and Eugenio Culurciello.
Linknet: Exploiting encoder representations for eﬃcient semantic
segmentation.
In 2017 IEEE Visual Communications and Image Processing (VCIP),
pages 1–4. IEEE, 2017.

References III
Vladimir Iglovikov and Alexey Shvets.
Ternausnet: U-net with VGG11 encoder pre-trained on imagenet for
image segmentation.
CoRR, abs/1801.05746, 2018.

Deep Learning based Segmentation Pipeline
for Label-Free Phase-Contrast Microscopy Images
Aydin Ayanzadeh, Özden Yal¸cın Özuysal, Devrim Pesen Okvur, Sevgi
Önal, Beh¸cet U˘gur Töreyin, Devrim Ünay
Istanbul Technical University, Izmir Institute of Technology,
Izmir Demokrasi University
September 23, 2020

Deep Learning based Segmentation Pipeline for Label-Free Phase-Contrast Microscopy Images

More Related Content

What's hot (20)

Similar to Deep Learning based Segmentation Pipeline for Label-Free Phase-Contrast Microscopy Images (20)

More from UMBC (20)

Recently uploaded (20)

Deep Learning based Segmentation Pipeline for Label-Free Phase-Contrast Microscopy Images