This study aimed to develop a local dataset of abnormal RBC morphology from confirmed cases of anemia and thalassemia in Thailand, providing a foundation for medical image analysis and future AI-assisted diagnostics. Blood smear samples from six hematological disorders were collected between April and May 2025, with twelve regions of interest segmented into approximately 34,000 single-cell images. To characterize cell variability, a convolutional autoencoder was applied to extract latent features, while ellipse fitting was used to quantify cell geometry. Expert hematologists validated representative clusters to ensure clinical accuracy, and data augmentation was employed to address class imbalance and expand rare morphological types. From the dataset, 14,089 high-quality single-cell images were used to classify RBC morphology into 36 clinically meaningful categories. Unlike existing datasets that rely on limited or curated samples, this dataset reflects population-specific characteristics and morphological diversity relevant to Southeast Asia. The results demonstrate the feasibility of establishing scalable and interpretable datasets that integrate computational methods with expert knowledge. The proposed dataset serves as a robust resource for advancing hematology research and contributes to bridging traditional diagnostics with AI-driven clinical support systems.

Red blood cell (RBC) morphology is critical for diagnosing hematological disorders, particularly in regions such as Southeast Asia where anemia and thalassemia are highly prevalent. However, manual microscopic assessment is labor-intensive, subjective, and dependent on expert availability, while existing automated methods often rely on small, curated datasets that fail to represent real-world smear variability. This study proposes a hybrid framework integrating preprocessing, unsupervised autoencoding, k-means clustering, ellipse fitting, expert-in-the-loop validation, and targeted data augmentation to establish a robust RBC labeling pipeline. High-resolution smear images from confirmed anemia and thalassemia cases were processed to extract over 14,000 single-cell patches, filtered systematically into quality-assured subsets. Latent features from a CNN autoencoder enabled clustering into 80 morphological groups, quantified using ellipse-based geometric metrics and validated by hematology experts. Data augmentation addressed class imbalance, expanding rare morphologies while preserving realistic cell structure. The resulting dataset captures clinically relevant morphological diversity specific to the Thai population and provides a scalable, interpretable framework for medical analysis and future AI model development in hematology.