After hematology training, participants will classify and/or annotate:
Neutrophils.
Lymphocytes.
Monocytes.
Eosinophils.
Basophils.
Immature granulocytes where applicable.
Erythroid cells where relevant.
Platelets.
Artifacts and difficult cells.
Expected outputs include:
Curated peripheral-smear cell dataset.
Cell-level annotations.
Standardized cell taxonomy.
Inter-rater agreement.
Candidate multiclass classifier.
Per-class sensitivity/precision/F1.
Confusion matrix and error analysis.
Manuscript, poster, certificates.
Artificial Intelligence in Healthcare
Artificial Intelligence, Hematology, Laboratory Medicine
Jorge Racedo at microscopIA, External Clinicians
Open internationally to medical students, health-science students, physicians, biomedical engineers, data-science students, laboratory professionals, and other students or professionals whose backgrounds are relevant to the project.
Prior experience in artificial intelligence is not required. Participants will receive project-specific training before independent annotation begins.
Applicants should have:
Professional or academic interest in the clinical area.
Sufficient English proficiency for scientific work.
Reliable internet access.
Strong attention to detail.
Ability to follow standardized research protocols.
Availability for approximately 3–5 hours per week.
Willingness to participate in scientific writing and quality-control activities.
Participation is offered through microscopIA membership. Selected applicants must hold a membership tier that permits enrollment in Join a Project and comply with the applicable Membership and Project Terms throughout their participation.
Any Country
Clinical Annotation, Data Curation, Data Validation, Data Quality Control, Literature Review, Database Development, Manuscript Preparation, Manuscript Revision, Poster Preparation, Conference Presentation, Response to Reviewers, AI Training (Data Processing and Labeling)
7
English, Spanish
External PubMed-indexed methodological precedent: Acevedo A, et al. Recognition of peripheral blood cell images using convolutional neural networks. Comput Methods Programs Biomed. 2019;180:105020. PMID: 31425939. The study used 17,092 expert-labeled images across eight peripheral blood-cell groups.
- Training will cover blood-cell morphology, classification taxonomy, abnormal/uncertain morphology, image quality, annotation consistency and hematologist adjudication.
15 de octubre de 2026 a las 0:00:00
20 de octubre de 2026
26 de octubre de 2026
15 de diciembre de 2026
Each participant may declare a maximum of two institutional affiliations.
Format:
Department, Organization (Hospital, University, Company), City, Country.microscopIA will cover approved publication fees associated with the anticipated regional manuscript.
Participants will not be required to personally pay approved article-processing charges.
Memberships are reinvested to pay microscopIA staff (epidemiologists, engineers, data scientists) salaries and to cover the publication fees for the resulting manuscripts. This maintain our work neutral, financially self-sustainable, and without any existing conflict of interest to maintain rigor and integrity in evidence generation.