Sign in Subscribe
Machine learning

Breakthrough in Artificial Intelligence: Nuclei Segmentation and Machine Learning

Strategic Insights

— 1 min read

Researchers from Inje University, in partnership with Princess Nourah Bint Abdulrahman University, have made significant advancements in artificial intelligence by employing machine learning approaches for nuclei segmentation in pathological images. This innovative work assesses the quality of nuclei image segmentation using various methods, including K-means clustering, Random Forest, Support Vector Machine, and Logistic Regression with Convolutional Neural Networks-derived features. The findings highlight the potential of leveraging CNN-based features in conjunction with Logistic Regression for enhancing the accuracy of cell nuclei segmentation.

Key Takeaways:

  • The study focused on nuclei segmentation, a crucial aspect in automated microscopic image analysis for early disease detection, including prostate cancer, breast cancer, brain tumors, and other diagnoses.
  • Researchers employed several machine learning methods, including K-means clustering, Random Forest, Support Vector Machine, and Logistic Regression with handcrafted and CNN-derived features.
  • The study achieved an accuracy of 96.90%, a Dice coefficient of 74.24, and a Jaccard coefficient of 55.61 using Logistic Regression based on CNN-derived features.
  • The results suggest that combining CNN-based features with Logistic Regression significantly enhances the accuracy of cell nuclei segmentation in pathological images.
  • In contrast, other methods like Random Forest, Support Vector Machine, and K-means algorithms yielded lower segmentation performance metrics.
  • The study's conclusions imply that the developed approach holds promise for refining computer-aided pathology workflows, leading to more reliable and earlier disease diagnoses.
  • The research was conducted by a team of authors from Inje University, including Rashadul Islam Sumon, Md Ariful Islam Mozumdar, Salma Akter, Shah Muhammad Imtiyaj Uddin, Mohammad Hassan Ali Al-Onaizan, Reem Ibrahim Alkanhel, and Mohammed Saleh Ali Muthanna.
  • Funders for this research include the Princess Nourah Bint Abdulrahman University.

Statistics:

  • The study achieved an accuracy of 96.90% using Logistic Regression based on CNN-derived features.
  • The Dice coefficient was found to be 74.24.
  • The Jaccard coefficient was found to be 55.61.

Sources:

  • Diagnostics. (2025,15(10),1271). Comparative Study of Cell Nuclei Segmentation Based on Computational and Handcrafted Features Using Machine Learning Algorithms.
  • MDPI AG. (No Publication Date). Diagnostics. (https://doi-org.sdpl.idm.oclc.org/10.3390/diagnostics15101271)
  • NewsRx. (2025, Jun 13). Reports from Inje University Add New Study Findings to Research in Machine Learning (Comparative Study of Cell Nuclei Segmentation Based on Computational and Handcrafted Features Using Machine Learning Algorithms). Health & Medicine Week. p 4570.

Read more