IRO Journals
Journal of Innovative Image Processing

Explainable Embryo Viability Classification Using Multi-Scale Residual Attention Networks

Open Access
Published: 06 May, 2026
Pages: 596-616
  • Marium Basheer Mahmood , Marium Basheer Mahmood
    Institute of Informatics for Postgraduate Studies, University of Information and Communications Technology, Baghdad, Iraq
    Institute of Informatics for Postgraduate Studies, University of Information and Communications Technology, Baghdad, Iraq

    Institute of Informatics for Postgraduate Studies, University of Information and Communications Technology, Baghdad, Iraq. 

    Department of Financial Affairs, University of Baghdad, Baghdad, Iraq

  • Ahmed Sabah Ahmed ALJumaili Ahmed Sabah Ahmed ALJumaili
    Department of Business Information Technology (BIT), College of Business Informatics, University of Information Technology and Communications, Iraq
    Department of Business Information Technology (BIT), College of Business Informatics, University of Information Technology and Communications, Iraq
view PDF
view PDF

How to Cite

Mahmood, Marium Basheer, and Ahmed Sabah Ahmed ALJumaili. 2026. “Explainable Embryo Viability Classification Using Multi-Scale Residual Attention Networks”. Journal of Innovative Image Processing 8 (2): 596-616. https://doi.org/10.36548/jiip.2026.2.009.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Embryo Viability Classification
In Vitro Fertilization
Convolutional Neural Networks
Multi-Scale Feature Fusion
Spatial Attention

Abstract

Embryo viability is essential for in vitro fertilization to achieve a successful transfer, implantation, and pregnancy outcome. However, automatic embryo classification remains a difficult task, as microscopy image collections are often limited in number, not balanced between classes, and challenging for clinical staff to read and trust completely. This paper presents a robust binary model for assessing the viability of embryos from images taken during IVF using a microscope. The proposed network is characterized by a hierarchical architecture with convolution, which consists of Conv Block modules, multi-scale residual blocks, depthwise separable convolutions, residual pathways, and spatial attention. The Hung Vuong Hospital Embryo Classification dataset was adopted, consisting of 801 cleaned and annotated images of embryos. Stratification and splitting were applied to the data to create training, validation, and test sets without any leakage. To balance the classes within the training dataset, only the minority class underwent data augmentation, while the other classes remained unchanged. The proposed algorithm was compared with EfficientNetB0, DenseNet121, and ResNet50 based on accuracy, precision, recall, F1 score, confusion matrix, ROC AUC, and PR AUC scores. The latter is particularly relevant when the data is not balanced. Additionally, the use of Grad-CAM heat maps and parameter randomization sanity tests has been applied to determine whether the generated explanations are grounded in consistent performance of the model or only seem reasonable visually. The proposed model achieved an accuracy of 0.96, precision of 0.91, recall of 0.94, F1 score of 0.92, ROC AUC of 0.92, and PR AUC of 0.90. Confusion matrix analysis revealed that the model obtained 11 true positives, 1 false negative, 2 false positives, and 67 true negatives. Consequently, it can be concluded that the proposed framework is capable of delivering accurate embryo viability classifications supported by images that can be interpreted in the context of medical imaging studies with insufficient amounts of data.

References

  1. Barhoun, Abbas, Mohammad Ali Balafar, Amin Golzari Oskouei, and Leila Sadeghi. "MDMBG-Net: A Multi-Task Deep Learning Model Addressing Class Imbalance for Blastocyst Grading in IVF." Biomedical Signal Processing and Control 116 (2026): 109455.
  2. Chen, Kuo, Jing Zuo, Wei Han, and Jin-hong Guo. "Intelligent Assisted Reproduction: Innovative Applications of Artificial Intelligence in Embryo Health Assessment." LabMed Discovery 2, no. 2 (2025): 100075.
  3. Kodali, Radha, Venkata Rao Dhulipalla, Venkata Siva Kishor Tatavarty, Madhavi Nadakuditi, Bharadwaj Thiruveedhula, Suryanarayana Gunnam, Durga Prasad Bavirisetti, and Gogulamudi Pradeep Reddy. "Interpretation of Deep Learning Model in Embryo Selection for in Vitro Fertilization (IVF) Treatment." arXiv preprint arXiv:2506.06680 (2025).
  4. Saraniya, M., and J. Anitha Ruth. "EmbryoNet-VGG16 Framework for Deep Learning-Based Embryo Classification with Otsu Segmentation." Discover Artificial Intelligence 5, no. 1 (2025): 194.
  5. Kishida, Takuma, and Haruki Nishizawa. "Usefulness of Embryo Evaluation via Artificial Intelligence–Based Image Analysis." Fujita Medical Journal 12, no. 1 (2026): 29-32.
  6. Borna, Mahdi-Reza, Mohammad Mehdi Sepehri, and Behnam Maleki. "An Artificial Intelligence Algorithm to Select Most Viable Embryos Considering Current Process in IVF Labs." Frontiers in artificial intelligence 7 (2024): 1375474.
  7. Saraniya, M., and J. Anitha Ruth. "Deep Learning-Based Embryo Quality Assessment: A Dual-Branch CNN Model Integrating Morphological and Spatial Features." Intelligence-Based Medicine 12 (2025): 100273.
  8. Boucret, Lisa, Floris Chabrun, Magalie Boguenet, Pascal Reynier, Pierre-Emmanuel Bouet, and Pascale May-Panloup. "Deep-Learning Model for Embryo Selection Using Time-Lapse Imaging of Matched High-Quality Embryos." Scientific reports 15, no. 1 (2025): 28068.
  9. Wu, C-Y., W-C. Wang, and P-K. Yang. "Application of Hyperspectral Imaging Technique and Artificial Intelligence on Quality Prediction of Embryonic Cells." Experimental Mechanics 66, no. 1 (2026): 207-220.
  10. Mienye, Ibomoiye Domor, Theo G. Swart, George Obaido, Matt Jordan, and Philip Ilono. "Deep Convolutional Neural Networks in Medical Image Analysis: A Review." Information 16, no. 3 (2025): 195.
  11. Raghu, Maithra, Chiyuan Zhang, Jon Kleinberg, and Samy Bengio. "Transfusion: Understanding Transfer Learning for Medical Imaging." Advances in neural information processing systems 32 (2019).
  12. Kartik, Kunal, Tafeer Ahmed, Shantanu Ghosh, and Rajat Gupta. "Transfer Learning with CNNs in Small ML Datasets: Applying Pre-Trained CNN Models and Fine-Tuning them for Limited Data Scenarios." Transfer 7, no. 5 (2023).
  13. Miled, Wided Souid, and Nozha Chakroun. "Leveraging Attention Mechanisms for Interpretable Human Embryo Image Segmentation." In ICAART (2), pp. 872-879. 2025.
  14. Arrieta, Alejandro Barredo, Natalia Díaz-Rodríguez, Javier Del Ser, Adrien Bennetot, Siham Tabik, Alberto Barbado, Salvador García et al. "Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges Toward Responsible AI." Information fusion 58 (2020): 82-115.
  15. Adebayo, Julius, Justin Gilmer, Michael Muelly, Ian Goodfellow, Moritz Hardt, and Been Kim. "Sanity Checks for Saliency Maps." Advances in neural information processing systems 31 (2018).
  16. Afnan, Michael Anis Mihdi, Yanhe Liu, Vincent Conitzer, Cynthia Rudin, Abhishek Mishra, Julian Savulescu, and Masoud Afnan. "Interpretable, not Black-Box, Artificial Intelligence Should be Used for Embryo Selection." Human reproduction open 2021, no. 4 (2021): hoab040.
  17. Abdulaal, Alaa Hussein, Ali H. Abdulwahhab, Aqeel Majeed Breesam, Zahra Hasan Oleiwi, Riyam Ali Yassin, Morteza Valizadeh, and Saja Nafea Mohsin. "MEMF-Net: A Mega-Ensemble of Multi-Feature CNNs for Classification of Breast Histopathological Images." Iraqi Journal for Computer Science and Mathematics 6, no. 3 (2025): 36.
  18. Movahedi, Faezeh, Rema Padman, and James F. Antaki. "Limitations of Receiver Operating Characteristic Curve on Imbalanced Data: Assist Device Mortality Risk Scores." The Journal of thoracic and cardiovascular surgery 165, no. 4 (2023): 1433-1442.
  19. Citarella, Alessia Auriemma, Pietro Battistoni, Chiara Coscarelli, Fabiola De Marco, Luigi Di Biasi, and Mengyuan Wang. "EmbryoVision AI: An Explainable Deep Learning Framework for Enhanced Blastocyst Selection in Assisted Reproductive Technologies." Image and Vision Computing (2025): 105795.
  20. Panaite, Doru-Răzvan, Vlad Barbu, Radu-Andrei Rosu, George Stoica, Şerban-Gabriel Doncean, and Mihaela Elena Breabăn. "Advanced Methods for Dealing with High Data Imbalance for Embryo Fertility Classification." Procedia Computer Science 246 (2024): 82-90.
  21. Zhang, Chensheng, Xintong Shi, Xinyue Yin, Jiayi Sun, Jianhui Zhao, and Yi Zhang. "Cleavage-Stage Embryo Segmentation Using SAM-Based Dual Branch Pipeline: Development and Evaluation with the CleavageEmbryo Dataset." Bioinformatics 41, no. 4 (2025): btae617.
  22. Ishaq, Muhammad, Salman Raza, Hunza Rehar, Shan E. Zain ul Abadeen, Dildar Hussain, Rizwan Ali Naqvi, and Seung-Won Lee. "Assisting the Human Embryo Viability Assessment by Deep Learning for in Vitro Fertilization." Mathematics 11, no. 9 (2023): 2023.
  23. Sarunpm. "Embryo Viability Prediction Dataset." Kaggle, 2023. https://www.kaggle.com/datasets/sarunpm/embryo-prediction.
  24. Kalatehjari, Maryam, Younes Ghasemi, Shaghayegh Mahmoudiandehkordi, Fatemeh Afrazeh, Hossein Abbasi, and Fariba Ghasemi. "Human Embryo Quality Assessment with Deep Learning Models: M. Kalatehjari et al." The Journal of Obstetrics and Gynecology of India 75, no. 3 (2025): 227-232