Skip to content
Open Imaging Index
CT MRI PET Liver

TCGA-LIHC

The Cancer Genome Atlas Liver Hepatocellular Carcinoma Collection

Clinical CT, MRI and PET of patients with liver cancer (hepatocellular carcinoma) whose tumors were profiled by The Cancer Genome Atlas, so imaging can be linked to genomics and outcome. 97 patients from four sites.

Go to dataset Edit Report

Overview

TCGA-LIHC is the radiology part of The Cancer Genome Atlas liver hepatocellular carcinoma project. The Cancer Imaging Archive hosts clinical CT, MRI and PET of TCGA patients, while their clinical, pathology and genomic data sit in the NCI Genomic Data Commons. Both use the same TCGA patient id, so imaging features can be linked to tumor genotype and patient outcome. The first version appeared in 2014, and version 5 followed in 2020.

Composition

Version 5 holds 97 patients with 237 studies, 1,688 DICOM series and 125,397 images (56.38 GB). Counted from the open TCIA series metadata, 75 patients have CT, 40 have MRI and one has a single whole-body PET series. There are 66 men and 31 women. Age at the earliest archived study ranges from 20 to 85 years, with a median of 64. Many patients have several studies: 56 of the 97 have more than one. The collection page also offers a snapshot of the TCGA clinical data from January 2016, plus the TCGA case forms that explain its fields.

Acquisition

Images come from routine care rather than a research protocol, contributed by Mayo Clinic, the University of North Carolina, Alberta Health Services and Lahey Hospital & Medical Center. Scanners are from Siemens, GE, Philips and Toshiba. Protocols differ between sites and studies: CT includes non-contrast and contrast-enhanced abdominal scans, and liver MRI series include T1-weighted gradient-echo before and after gadolinium, T2-weighted fast spin-echo, in- and opposed-phase imaging and, for a few patients, MR elastography. The baseline studies of each case were acquired before surgery. DICOM dates are shifted by a random per-site offset, which keeps intervals between studies intact.

Annotations

The collection itself contains no image labels. Tumor and organ segmentations made later by other groups are published as separate analysis datasets.

Known limitations

  • Scanner types, protocols and series naming vary widely, and no harmonized sequence labels are provided.
  • Only 97 of the TCGA liver cancer patients have imaging, so the imaged subset may not represent the full TCGA cohort.
  • TCIA and TCGA use different date conventions that are not reconciled.
  • Two patients have series without a manufacturer value.

Cohort

Aggregate numbers from the sources below. Bars are relative to the 97 subjects.

Sex

  • Female 31 32%
  • Male 66 68%

PatientSex in TCIA NBIA API getPatient metadata for TCGA-LIHC (open, no account, CC BY 3.0), PatientSex field

Modality

  • CT 75 77%
  • MRI 40 41%
  • PET 1 1%

Modality in TCIA NBIA API getSeries metadata for TCGA-LIHC (open, no account, CC BY 3.0), counted per patient id from the Modality and Manufacturer fields

Age

mean 61.4 ± 13.5, median 64, range 20 to 85

0
10
20
30
37
20-2930-3940-4950-5960-6970-7980-89

PatientAge in TCIA NBIA API getPatientStudy metadata for TCGA-LIHC (open, no account, CC BY 3.0), PatientAge of each patient's earliest study

Scanner vendor

  • Siemens Healthineers 60 62%
  • GE HealthCare 55 57%
  • Philips 4 4%
  • Canon Medical 3 3%

Manufacturer in TCIA NBIA API getSeries metadata for TCGA-LIHC (open, no account, CC BY 3.0), counted per patient id from the Modality and Manufacturer fields

License and access

Our reading of the license, not legal advice. Before you use the data, read the original license and confirm that your use is allowed. We take no responsibility for how you use a dataset. Full disclaimer

Access
Open download

Download without an account

DICOM images via the TCIA Data Retriever. Clinical, genomic and histopathology data of the same patients are in the NCI Genomic Data Commons; a clinical snapshot is an open download from the collection page.

Access page

Creative Commons Attribution 3.0 Unported

Older version of CC BY. Use, share and adapt the data for any purpose, including commercial use, as long as you credit the creators.

TCIA lists the images under CC BY 3.0. Users must also follow the TCIA Data Usage Policy, which says users must "Not use the requested datasets, either alone or in concert with any other information, to identify or contact individual participants" and must cite the dataset DOI (https://www.cancerimagingarchive.net/data-usage-policies-and-restrictions/). Publications are asked to acknowledge the TCGA Research Network.

Original license text Version read: 3.0 Checked 2026-10-07

What you can do

  • Yes
  • Yes
  • Yes
  • Yes
  • Yes

What you can share

  • Yes
  • Yes
  • Yes

What you must do

  • Yes
  • Share alike No
  • No
  • No
  • Manuscript review No
  • Release code No
  • Return results No
  • Delete after use No

Limits

  • No
  • Location limits No

Citation

Erickson BJ, Kirk S, Lee Y, Bathe O, Kearns M, Gerdes C, Rieger-Christ K, Lemmerman J. (2016). The Cancer Genome Atlas Liver Hepatocellular Carcinoma Collection (TCGA-LIHC) (Version 5) [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/K9/TCIA.2016.IMMQW8UQ

Sources

Every number on this page comes from one of these documents. Each chart names the table or page it is taken from. The raw numbers are in stats.csv.