---
license: cc
dataset_info:
  features:
  - name: image_id
    dtype: int64
  - name: image
    dtype: image
  - name: width
    dtype: int64
  - name: height
    dtype: int64
  - name: objects
    struct:
    - name: area
      sequence: float64
    - name: bbox
      sequence:
        sequence: int64
    - name: category
      sequence: int64
    - name: id
      sequence: int64
  splits:
  - name: train
    num_bytes: 181919726.0
    num_examples: 547
  download_size: 181637597
  dataset_size: 181919726.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

# breastcanc-ultrasound-class

## Background

Cancer is the second leading cause of death worldwide, according to _IHME - Global Burden of Disease_, with 10.7 mln casualties in 2019. 

![total_deaths](./annual-number-of-deaths-by-cause.png)

Amongst the various types of cancer, a huge role is played by breast cancer, which stands in 4th position among the deadliest tumors, with more than 700.000 deaths during 2019 (_IHME - Global Burden of Disease_).

![death_composition](./total-cancer-deaths-by-type.png)

Moreover, breast cancer has the highest share of number of cases/100 people worldwide (0.23 cases/100 people; _IHME - Global Burden of Disease_), as shown in [table1](#table-1)
:

### Table 1

| Type of Cancer                       | Cases per 100 people                      |
|--------------------------------------|-------------------------------------------|
| Breast Cancer                        | 0.23                                     |
| Colon and Rectum Cancer              | 0.14                                      |
| Prostate cancer                      | 0.13                                      |
| Bladder Cancer                       | 0.034                                     |
| Stomach Cancer                       | 0.033                                     |

In this sense, it is more than vital to put intense effort into precision medicine and diagnostic tools for what concerns breast cancer: part of this effort should involve making curated dataset of diagnostic images available to the large public.

## Dataset source and composition

This dataset is part of the one kindly provided by [Walid Al-Dhabyani and collaborators in 2019](https://doi.org/10.1016/j.dib.2019.104863), and encompasses 547 images, 387 representing benign breast cancer ultrasound images and 160 of them representing malignant breast cancer ultrasound images, with their related object detection information (in the same format of [CPPE-5](https://huggingface.co/datasets/cppe-5)) inferred from masks.

## License, references and citation 

The dataset is hereby provided under CC family licenses. 

Please cite Al-Dhabyani W, Gomaa M, Khaled H, Fahmy A. Dataset of breast ultrasound images. Data in Brief. 2020 Feb;28:104863. DOI: 10.1016/j.dib.2019.104863 when using it.