Histology Image Collection Library

Welcome to the Histology Image Collection Library

The HICL is a complilation of 3870 histopathological images from various diseases, such as brain cancer, breast cancer and HPV (Human Papilloma Virus)-Cervical cancer.
The number of images for each disease type is summarized below:

Breast cancer: Excluding skin cancer, breast cancer is the most common cancer among women, accounting for nearly 1 in 3 cancers diagnosed in US women. While clinical assessment clues (breast examination or imaging results) may be strongly suggestive of a cancer diagnosis, microscopic analysis of breast tissue is necessary for a definitive diagnosis of breast cancer and to determine whether the cancer is in situ or invasive. Our breast cancer data comprise 116 breast cancer cases. In total, 872 images were generated, among which 414 were IHC stained (x40) and 458 were H&E stained (227 at x20 and 231 x40). Two light microscopy imaging systems were used for digitization, the first comprised a Zeiss Axiostar-Plus (Zeiss, Göttingen, Germany) microscope connected to a LEICA DC 300 F (Leica Microsystems GmbH) camera and the second consisted of a LEICA DM 2500 microscope connected to a LEICA DFC 420C camera (Leica Microsystems GmbH). The raw clinical material was collected from the archives of the University Hospital of Patras, Greece.
Kostopoulos, S., Glotsos, D., Cavouras, D., Daskalakis, A., Kalatzis, I., Georgiadis, P., Bougioukos, P., Ravazoula, P., Nikiforidis, G.Computer-based association of the texture of expressed estrogen receptor nuclei with histologic grade using immunohistochemically-stained breast carcinomas (2009) Analytical and Quantitative Cytology and Histology, 31 (4), pp. 187-196.

Brain cancer: Brain tumours are considered one of the most lethal and difficult to treat forms of cancer, with unknown aetiology and lack of any realistic screening. In the diagnosis of astrocytomas, grade characterization is of major clinical importance, since it provides an index of disease severity and determinatively influences patient treatment. Our brain cancer data comprise 93 brain cancer cases. In total, 2548 H&E stained images of different grades were generated at x20 and x40, among which 840 were created from low grade cases, 1608 from high grade cases and 100 from ambiguous (low to high grade) cases. The light microscopy imaging system used for digitization comprised a LEICA DM 2500 microscope connected to a LEICA DFC 420C camera (Leica Microsystems GmbH). The raw clinical material was collected from the archives of the University Hospital of Patras, Greece.
Glotsos, D., Kalatzis, I., Spyridonos, P., Kostopoulos, S., Daskalakis, A., Athanasiadis, E., Ravazoula, P., Nikiforidis, G., Cavouras, D. Improving accuracy in astrocytomas grading by integrating a robust least squares mapping driven support vector machine classifier into a two level grade classification scheme (2008) Computer Methods and Programs in Biomedicine, 90 (3), pp. 251-261.

Laryngeal cancer: Laryngeal cancer is the most frequent cancer among head and neck neoplasms. It amounts to about 3 % of newly diagnosed cancers and has poor prognosis. A new prognostic factor for laryngeal cancer is the production of p63 protein by the TP63 gene and p63 over-expression in epithelial neoplasms of the head and neck squamous cell carcinomas. Our laryngeal cancer data comprise 55 laryngeal cancer cases. In total, 450 P63 stained images of different grades were generated at x20 and x40, among which 175 were created from grade I cases, 147 from grade II cases and 128 from grade III cases. The light microscopy imaging system used for digitization comprised a LEICA DM 2500 microscope connected to a LEICA DFC 420C camera (Leica Microsystems GmbH). The raw clinical material was collected from the archives of the University Hospital of Patras, Greece.
Ninos, K., Kostopoulos, S., Kalatzis, I., Sidiropoulos, K., Ravazoula, P., Sakellaropoulos, G., Panayiotakis, G., Economou, G., Cavouras, D.,Microscopy image analysis of p63 immunohistochemically stained laryngeal cancer lesions for predicting patient 5-year survival (2016) European Archives of Oto-Rhino-Laryngology, 273 (1), pp. 159-168.

All images were collected from the University Hospital of Patras, Greece.
All images were assigned to a gold standard diagnosis based on the assessment of expert physicians (histopathologists).
Each image was obtained from the most representative parts of the tumor. This region was marked by expert physicians (histopathologists).


Each image has the following associated information:
ID: a unique identification number is assigned to each image
Magnification factor: images have been digitized under different magnification factors (i.e. 5x, 10x, 20x, 40x, 63x)
Staining: images have been processed with different stains (i.e. Haematoxylin & Eosin (H&E), Immunohistochemistry ( IHC))
Diagnosis: Depending on the diseases type, images have been assigned to a gold standard diagnosis (i.e. grade I, II, III, low grade, high grade, normal, abnormal, benign, cancer etc)
Additional information: Some images may contain additional explanatory information (i.e. qualitative assessments by expert physician, such as presence, absence or marked necrosis, mild, medium or marked cellularity etc)

The above information is included in the title of each image. An example of decoding the title of each image is the following:

IMAGE: 39_HE_40X_LEICA_GIII_PATRA_BRAIN_CANCER_1.tif

39: Patient ID
HE: Staining Protocol
40X: Magnification Factor
LEICA: Microscope Used
GIII: Histological Diagnosis - Grade III  
PATRA: Hospital-City
BRAIN_CANCER: Disease
1: The Number of Image at the Specific ID

License agreement
The HICL is available for use under an academic license in the field of computer science and medical imaging for timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by MEDISP.

Permitted Use
The license agreement permits the following:

1. The collection can be used only for academic purposes in the field of computer science. The collection cannot be used for commercial and teaching purposes.
2. In case of publication to scientific journals, conferences etc, a limited number of images may be included in the paper (up to 10).
3. The collection shall not be made available for download or access by any other third party than the Licensee.
4. In case of publications, the following references should be made:


• Breast cancer: Kostopoulos, S., Glotsos, D., Cavouras, D., Daskalakis, A., Kalatzis, I., Georgiadis, P., Bougioukos, P., Ravazoula, P., Nikiforidis, G.Computer-based association of the texture of expressed estrogen receptor nuclei with histologic grade using immunohistochemically-stained breast carcinomas (2009) Analytical and Quantitative Cytology and Histology, 31 (4), pp. 187-196.
• Brain cancer: Glotsos, D., Kalatzis, I., Spyridonos, P., Kostopoulos, S., Daskalakis, A., Athanasiadis, E., Ravazoula, P., Nikiforidis, G., Cavouras, D.Improving accuracy in astrocytomas grading by integrating a robust least squares mapping driven support vector machine classifier into a two level grade classification scheme (2008) Computer Methods and Programs in Biomedicine, 90 (3), pp. 251-261.
• Laryngeal cancer: Ninos, K., Kostopoulos, S., Kalatzis, I., Sidiropoulos, K., Ravazoula, P., Sakellaropoulos, G., Panayiotakis, G., Economou, G., Cavouras, D.,Microscopy image analysis of p63 immunohistochemically stained laryngeal cancer lesions for predicting patient 5-year survival (2016) European Archives of Oto-Rhino-Laryngology, 273 (1), pp. 159-168.


5. The Licensee is responsible for the proper or improper use of the data. MEDISP is not responsible for any improper use of the HICL.