MétaCan
Menu
← Back to cohort
Record W6902586398 · doi:10.7910/dvn/aqepin

BOSQUE Test set

2025· dataset· en· W6902586398 on OpenAlexaff

Bibliographic record

VenuePolyPublie (École Polytechnique de Montréal) · 2025
Typedataset
Languageen
Field
Topic
Canadian institutionsPolytechnique Montréal
Fundersnot available
KeywordsMetadataTest (biology)Test setSkin lesionData setSet (abstract data type)Focus (optics)

Abstract

fetched live from OpenAlex

BOSQUE Test Set: A Dermoscopic Image Dataset from Colombian Patients with Diverse Skin Phototypes Description: The BOSQUE Test Set is a curated dataset of 151 dermoscopic images of pigmented skin lesions, collected from dermatology consultations and outreach campaigns in Bogotá, Colombia. Each image is accompanied by expert-verified metadata including histological diagnosis, patient demographic details, anatomical site, and skin phototype. The dataset is intended to support machine learning research in dermatology with a particular focus on skin tone diversity and fairness in diagnostic algorithms. The dataset was developed under the guidance of Universidad El Bosque, whose name inspired the acronym BOSQUE. It responds to the global underrepresentation of darker skin phototypes in existing dermoscopic image collections such as HAM10000, and aims to improve diagnostic equity through inclusive data curation. Key Features 151 dermoscopic images acquired in real-world clinical settings Captured using polarized light dermatoscopes (DermLite 4 + iPhone) Inclusive population: Sex: 97 Female, 54 Male Age groups: from 0–29 to 90+, categorized into clinically relevant bins Fitzpatrick skin phototypes: ranging from II to VI Type II (fair, burns easily): 11 patients Type III (light brown, mild burns): 94 patients Type IV (moderate brown, rarely burns): 34 patients Type V (dark brown, very rarely burns): 7 patients Type VI (deeply pigmented, never burns): 5 patients Lesion characteristics: Nature: benign or malignant (histopathologically confirmed) Size: categorized as ≤5mm, 6–10mm, 11–20mm, >20mm Evolution time: grouped into <1y, 1y, 2y, 3–4y, 5–9y, and 10y+ categories Anatomical site: head/neck, trunk, limbs, or acral areas Histopathological diagnosis: 7-class ISIC-style labels (akiec, bcc, bkl, df, mel, nv, vasc) Clinical label: melanocytic vs. non-melanocytic (from clinical diagnosis) Clinical context: includes personal history of NMSC and use of photosensitizing drugs Image naming: pseudonymized file names encode diagnosis label and image ID Ethics: all data anonymized and collected under IRB-approved protocol in Colombia Included Files BOSQUE_test_set.zip: Folder containing 151 dermoscopic image files (JPG) BOSQUE_metadata.csv: Metadata for each image, including: Patient sex, age group, skin phototype Anatomical site of the lesion Lesion nature (benign/malignant) Lesion size and evolution time (binned) Histological diagnosis (7-class) Clinical label (melanocytic / non-melanocytic) Use Cases This dataset is intended for: Benchmarking AI models for dermoscopic image classification Fairness analysis across skin tones, sex, and age groups Medical education and clinical training on diverse skin phototypes Comparison against HAM10000 or ISIC datasets in research Ethical Statement All patients provided informed consent for the capture and use of clinical and dermoscopic images, the collection of relevant clinical metadata, and the performance of skin biopsies for diagnostic confirmation. The study protocol was reviewed and approved by the Institutional Ethics Committee at Subred Integrada de Servicios de Salud Norte E.S.E and Universidad El Bosque (Bogotá, Colombia). All data were anonymized in compliance with Colombian health data privacy regulations and international ethical standards (e.g., Declaration of Helsinki). No personally identifiable information is included in the metadata or image files. Access to data was restricted to authorized investigators, and patients were informed about the research and educational use of their anonymized data. Suggested Citation [Author(s)]. (2025). BOSQUE Test Set: A Dermoscopic Image Dataset from Colombian Patients with Diverse Skin Phototypes [Data set]. Harvard Dataverse. https://doi.org/xxxxx

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.013
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Not applicable · Consensus signal: Not applicable
GenreCandidate signal: Dataset · Consensus signal: Dataset
Teacher disagreement score0.111
Threshold uncertainty score0.371

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.013
Meta-epidemiology (narrow)0.0050.001
Meta-epidemiology (broad)0.0030.003
Bibliometrics0.0050.003
Science and technology studies0.0030.001
Scholarly communication0.0040.002
Open science0.0070.003
Research integrity0.0050.004
Insufficient payload (model declined to judge)0.1110.118

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.011
GPT teacher head0.255
Teacher spread0.244 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designNot applicable
Domainnot available
GenreDataset

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venuePolyPublie (École Polytechnique de Montréal)→French-language works237,207→