All active requirements
ActiveREQ-000022

Oncology Imaging & Clinical Data — Breast Cancer and NSCLC

Seeking eligible, de-identified clinical imaging datasets for two independent patient cohorts: approximately 1,000 patients with confirmed breast cancer and approximately 1,000 patients with confirmed non-small cell lung cancer (NSCLC). Imaging must be linked to the relevant radiology, pathology and molecular documentation specified below. Volume targets are indicative and subject to availability and eligibility; partial modality coverage per patient is acceptable as specified. Baseline/pre-treatment imaging is preferred; post-operative or post-treatment scans must be flagged.

Data type
Oncology DICOM imaging with linked clinical, pathology and molecular records
Volume
Breast cancer cohort: approximately 1,000 confirmed patientsNSCLC cohort: approximately 1,000 confirmed patients, independent of the breast cancer cohortVolume targets are indicative and subject to availability and eligibilityConfirm eligible patient counts by disease cohort, modality and supporting clinical-document availability
Delivery date
Not specified

Requirement brief

Specific requirements

  • Breast cancer — cohort and timing: Approximately 1,000 patients with confirmed breast cancer. Baseline/pre-treatment imaging is preferred; flag post-operative or post-treatment scans.
  • Breast cancer — mammography / DBT: Original DICOM images. Identify 2D mammography versus digital breast tomosynthesis (DBT); retain views and laterality.
  • Breast cancer — ultrasound: Original DICOM images and cine loops, where available.
  • Breast cancer — MRI: Complete relevant DICOM series, including contrast-enhanced sequences where available.
  • Breast cancer — FDG PET/CT: Original PET and corresponding CT DICOM series, where available. Retain tracer, uptake time and SUV details.
  • Breast cancer — FAPI PET/CT: Original PET and corresponding CT DICOM series, where available.
  • Breast cancer — imaging reports: Final report for every submitted imaging examination, matched to the correct DICOM study.
  • Breast cancer — histopathology: Final biopsy and/or surgical pathology report confirming breast cancer; include tumour type and grade where reported.
  • Breast cancer — IHC / biomarkers: ER, PR, HER2 and Ki-67; include molecular subtype and nodal status where available. Flag missing results.
  • NSCLC — cohort and timing: Approximately 1,000 patients with confirmed non-small cell lung cancer. Baseline/pre-treatment scans are preferred; flag scans obtained after surgery or cancer treatment.
  • NSCLC — chest CT: Complete original chest CT DICOM study, including all relevant series. Thin-section images are preferred.
  • NSCLC — FDG PET/CT: Original PET DICOM and corresponding CT DICOM from the same examination, where available.
  • NSCLC — availability mapping: For each patient, indicate chest CT, FDG PET/CT, or both.
  • NSCLC — imaging reports: Final report for every submitted CT or PET/CT examination, matched to its DICOM study.
  • NSCLC — histopathology: Final biopsy and/or surgical pathology report confirming NSCLC and subtype where reported.
  • NSCLC — EGFR molecular test: Final EGFR result with the specific mutation/variant where reported. This result is required for the primary EGFR prediction analysis.
  • Shared delivery — original imaging: Native/original DICOM files for each available and submitted modality, including all relevant views and series.
  • Shared delivery — linked records: Use the same de-identified patient ID across imaging, pathology, reports and biomarkers, with a unique examination/study ID for each study.
  • Shared delivery — treatment timing: Identify baseline/pre-treatment versus post-operative/post-treatment scans. Retain the examination date or relative timepoint without revealing patient identity.
  • Shared governance — privacy and licensing: De-identify images, DICOM metadata and clinical documents before sharing. Ensure authority to license and share the dataset for the intended AI/research use.
  • Shared delivery — availability: List patient counts by disease cohort, modality and supporting clinical-document availability. Clearly flag missing tests/results.
  • NSCLC metadata spreadsheet — patient/study IDs: Include the de-identified patient ID and unique examination/study ID.
  • NSCLC metadata spreadsheet — image mapping: Include modality, DICOM folder/filename, examination date or relative timepoint, and reconstruction series.
  • NSCLC metadata spreadsheet — diagnosis and molecular results: Include histopathology diagnosis/subtype, EGFR result and mutation, and other molecular results where available.
  • NSCLC metadata spreadsheet — technical details: Include scanner manufacturer/model, CT slice thickness and reconstruction parameters.
  • Supplier response: Confirm available eligible case counts for each cohort, modality coverage, availability of mandatory reports and molecular markers, and any known limitations.
  • Samples: Supply samples only through approved, secure channels after confirming appropriate data rights and de-identification.

Coverage

Required coverage

  • Two independent patient cohorts: confirmed breast cancer and confirmed non-small cell lung cancer (NSCLC)
  • Breast imaging: 2D mammography / DBT, ultrasound, breast MRI, FDG PET/CT and FAPI PET/CT, with availability as specified
  • NSCLC imaging: chest CT, FDG PET/CT, or both, mapped per patient
  • Matched imaging reports, histopathology and cohort-specific biomarker / molecular documentation
  • Partial modality coverage per patient is acceptable as specified; missing tests/results and limitations must be flagged

Interested in this requirement?

Review the supplier process and prepare your capabilities.

Be a supplier
Oncology Imaging & Clinical Data — Breast Cancer and NSCLC | Active Requirement