{"id":27147777,"url":"https://github.com/soroushj/mhsma-dataset","last_synced_at":"2025-07-21T22:33:08.095Z","repository":{"id":56160796,"uuid":"184467941","full_name":"soroushj/mhsma-dataset","owner":"soroushj","description":"MHSMA: The Modified Human Sperm Morphology Analysis Dataset","archived":false,"fork":false,"pushed_at":"2024-03-23T07:29:24.000Z","size":13746,"stargazers_count":31,"open_issues_count":0,"forks_count":10,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-04-08T11:51:34.138Z","etag":null,"topics":["dataset","deep-learning","machine-learning","medical-imaging"],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/soroushj.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-05-01T19:13:08.000Z","updated_at":"2025-04-01T07:23:28.000Z","dependencies_parsed_at":"2025-04-08T12:01:26.829Z","dependency_job_id":null,"html_url":"https://github.com/soroushj/mhsma-dataset","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/soroushj/mhsma-dataset","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/soroushj%2Fmhsma-dataset","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/soroushj%2Fmhsma-dataset/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/soroushj%2Fmhsma-dataset/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/soroushj%2Fmhsma-dataset/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/soroushj","download_url":"https://codeload.github.com/soroushj/mhsma-dataset/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/soroushj%2Fmhsma-dataset/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266392944,"owners_count":23922266,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-07-21T11:47:31.412Z","response_time":64,"last_error":null,"robots_txt_status":null,"robots_txt_updated_at":null,"robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["dataset","deep-learning","machine-learning","medical-imaging"],"created_at":"2025-04-08T11:51:08.939Z","updated_at":"2025-07-21T22:33:08.027Z","avatar_url":"https://github.com/soroushj.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# MHSMA: The Modified Human Sperm Morphology Analysis Dataset\n\nThe MHSMA dataset is a collection of human sperm images from 235 patients with male factor infertility.\nEach image is labeled by experts for normal or abnormal sperm acrosome, head, vacuole, and tail.\n\nThe training, validation, and test sets contain 1000, 240, and 300 images, respectively.\n\nImages are available in two different crop sizes: 128x128- and 64x64-pixel.\nThe following figure shows two versions of the same instance.\n\n|                  128x128-pixel                   |                  64x64-pixel                   |\n| :----------------------------------------------: | :--------------------------------------------: |\n| ![MHSMA-128 sample](sample/mhsma-128-sample.png) | ![MHSMA-64 sample](sample/mhsma-64-sample.png) |\n\nIn MHSMA, each instance is a grayscale image capturing a single sperm.\nThe head of the sperm is roughly located at the center of the image.\nAlso, the sperm tail is not entirely visible in the images.\n\nLabels can be either `0` (normal, positive) or `1` (abnormal, negative).\n\nThe dataset is available in `.npy` format.\nYou can load the `.npy` files using [numpy.load](https://docs.scipy.org/doc/numpy/reference/generated/numpy.load.html).\nThe details of the files are described in the table below.\n\n| File                   | Shape              | Type    | Description                           |\n| ---------------------- | ------------------ | ------- | ------------------------------------- |\n| `x_128_train.npy`      | `(1000, 128, 128)` | `uint8` | Training set, 128x128-pixel version   |\n| `x_128_valid.npy`      | `(240, 128, 128)`  | `uint8` | Validation set, 128x128-pixel version |\n| `x_128_test.npy`       | `(300, 128, 128)`  | `uint8` | Test set, 128x128-pixel version       |\n| `x_64_train.npy`       | `(1000, 64, 64)`   | `uint8` | Training set, 64x64-pixel version     |\n| `x_64_valid.npy`       | `(240, 64, 64)`    | `uint8` | Validation set, 64x64-pixel version   |\n| `x_64_test.npy`        | `(300, 64, 64)`    | `uint8` | Test set, 64x64-pixel version         |\n| `y_acrosome_train.npy` | `(1000,)`          | `uint8` | Training set labels for acrosome      |\n| `y_acrosome_valid.npy` | `(240,)`           | `uint8` | Validation set labels for acrosome    |\n| `y_acrosome_test.npy`  | `(300,)`           | `uint8` | Test set labels for acrosome          |\n| `y_head_train.npy`     | `(1000,)`          | `uint8` | Training set labels for head          |\n| `y_head_valid.npy`     | `(240,)`           | `uint8` | Validation set labels for head        |\n| `y_head_test.npy`      | `(300,)`           | `uint8` | Test set labels for head              |\n| `y_vacuole_train.npy`  | `(1000,)`          | `uint8` | Training set labels for vacuole       |\n| `y_vacuole_valid.npy`  | `(240,)`           | `uint8` | Validation set labels for vacuole     |\n| `y_vacuole_test.npy`   | `(300,)`           | `uint8` | Test set labels for vacuole           |\n| `y_tail_train.npy`     | `(1000,)`          | `uint8` | Training set labels for tail          |\n| `y_tail_valid.npy`     | `(240,)`           | `uint8` | Validation set labels for tail        |\n| `y_tail_test.npy`      | `(300,)`           | `uint8` | Test set labels for tail              |\n\nThe following table shows the number of positive and negative examples in the dataset.\n\n\u003ctable\u003e\n  \u003cthead\u003e\n    \u003ctr\u003e\n      \u003cth\u003eSet\u003c/th\u003e\n      \u003cth\u003eLabel\u003c/th\u003e\n      \u003cth\u003e# Positive\u003c/th\u003e\n      \u003cth\u003e# Negative\u003c/th\u003e\n      \u003cth\u003e% Positive\u003c/th\u003e\n    \u003c/tr\u003e\n  \u003c/thead\u003e\n  \u003ctbody\u003e\n    \u003ctr\u003e\n      \u003ctd rowspan=\"4\"\u003eWhole dataset\u003c/td\u003e\n      \u003ctd\u003eAcrosome\u003c/td\u003e\n      \u003ctd\u003e1,086\u003c/td\u003e\n      \u003ctd\u003e454\u003c/td\u003e\n      \u003ctd\u003e70.52\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eHead\u003c/td\u003e\n      \u003ctd\u003e1,122\u003c/td\u003e\n      \u003ctd\u003e418\u003c/td\u003e\n      \u003ctd\u003e72.86\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eVacuole\u003c/td\u003e\n      \u003ctd\u003e1,301\u003c/td\u003e\n      \u003ctd\u003e239\u003c/td\u003e\n      \u003ctd\u003e84.48\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eTail\u003c/td\u003e\n      \u003ctd\u003e1,471\u003c/td\u003e\n      \u003ctd\u003e69\u003c/td\u003e\n      \u003ctd\u003e95.52\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd rowspan=\"4\"\u003eTraining set\u003c/td\u003e\n      \u003ctd\u003eAcrosome\u003c/td\u003e\n      \u003ctd\u003e699\u003c/td\u003e\n      \u003ctd\u003e301\u003c/td\u003e\n      \u003ctd\u003e69.90\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eHead\u003c/td\u003e\n      \u003ctd\u003e727\u003c/td\u003e\n      \u003ctd\u003e273\u003c/td\u003e\n      \u003ctd\u003e72.70\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eVacuole\u003c/td\u003e\n      \u003ctd\u003e830\u003c/td\u003e\n      \u003ctd\u003e170\u003c/td\u003e\n      \u003ctd\u003e83.00\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eTail\u003c/td\u003e\n      \u003ctd\u003e954\u003c/td\u003e\n      \u003ctd\u003e46\u003c/td\u003e\n      \u003ctd\u003e95.40\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd rowspan=\"4\"\u003eValidation set\u003c/td\u003e\n      \u003ctd\u003eAcrosome\u003c/td\u003e\n      \u003ctd\u003e174\u003c/td\u003e\n      \u003ctd\u003e66\u003c/td\u003e\n      \u003ctd\u003e72.50\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eHead\u003c/td\u003e\n      \u003ctd\u003e176\u003c/td\u003e\n      \u003ctd\u003e64\u003c/td\u003e\n      \u003ctd\u003e73.33\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eVacuole\u003c/td\u003e\n      \u003ctd\u003e209\u003c/td\u003e\n      \u003ctd\u003e31\u003c/td\u003e\n      \u003ctd\u003e87.08\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eTail\u003c/td\u003e\n      \u003ctd\u003e233\u003c/td\u003e\n      \u003ctd\u003e7\u003c/td\u003e\n      \u003ctd\u003e97.08\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd rowspan=\"4\"\u003eTest set\u003c/td\u003e\n      \u003ctd\u003eAcrosome\u003c/td\u003e\n      \u003ctd\u003e213\u003c/td\u003e\n      \u003ctd\u003e87\u003c/td\u003e\n      \u003ctd\u003e71.00\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eHead\u003c/td\u003e\n      \u003ctd\u003e219\u003c/td\u003e\n      \u003ctd\u003e81\u003c/td\u003e\n      \u003ctd\u003e73.00\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eVacuole\u003c/td\u003e\n      \u003ctd\u003e262\u003c/td\u003e\n      \u003ctd\u003e38\u003c/td\u003e\n      \u003ctd\u003e87.33\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eTail\u003c/td\u003e\n      \u003ctd\u003e284\u003c/td\u003e\n      \u003ctd\u003e16\u003c/td\u003e\n      \u003ctd\u003e94.67\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/tbody\u003e\n\u003c/table\u003e\n\n## Results\n\nIf you would like to add a new result, you can [open a pull request](https://github.com/soroushj/mhsma-dataset/pulls).\n\n\u003ctable\u003e\n  \u003cthead\u003e\n    \u003ctr\u003e\n      \u003cth\u003eMethod\u003c/th\u003e\n      \u003cth\u003eLabel\u003c/th\u003e\n      \u003cth\u003eAccuracy\u003c/th\u003e\n      \u003cth\u003ePrecision\u003c/th\u003e\n      \u003cth\u003eRecall\u003c/th\u003e\n      \u003cth\u003eF\u003csub\u003e0.5\u003c/sub\u003e score\u003c/th\u003e\n      \u003cth\u003eG-mean\u003c/th\u003e\n      \u003cth\u003eROC AUC\u003c/th\u003e\n      \u003cth\u003eMCC\u003c/th\u003e\n    \u003c/tr\u003e\n  \u003c/thead\u003e\n  \u003ctbody\u003e\n    \u003ctr\u003e\n      \u003ctd rowspan=\"3\"\u003e\u003ca href=\"https://doi.org/10.1016/j.compbiomed.2019.04.030\"\u003eA novel deep learning method for automatic assessment of human sperm images\u003c/a\u003e (Apr 2019)\u003c/td\u003e\n      \u003ctd\u003eAcrosome\u003c/td\u003e\n      \u003ctd\u003e76.67\u003c/td\u003e\n      \u003ctd\u003e85.93\u003c/td\u003e\n      \u003ctd\u003e80.28\u003c/td\u003e\n      \u003ctd\u003e84.74\u003c/td\u003e\n      \u003ctd\u003e83.06\u003c/td\u003e\n      \u003ctd\u003e83.89\u003c/td\u003e\n      \u003ctd\u003e+0.4618\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eHead\u003c/td\u003e\n      \u003ctd\u003e77.00\u003c/td\u003e\n      \u003ctd\u003e83.48\u003c/td\u003e\n      \u003ctd\u003e85.39\u003c/td\u003e\n      \u003ctd\u003e83.86\u003c/td\u003e\n      \u003ctd\u003e84.43\u003c/td\u003e\n      \u003ctd\u003e77.80\u003c/td\u003e\n      \u003ctd\u003e+0.4053\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eVacuole\u003c/td\u003e\n      \u003ctd\u003e91.33\u003c/td\u003e\n      \u003ctd\u003e94.36\u003c/td\u003e\n      \u003ctd\u003e95.80\u003c/td\u003e\n      \u003ctd\u003e94.65\u003c/td\u003e\n      \u003ctd\u003e95.08\u003c/td\u003e\n      \u003ctd\u003e88.08\u003c/td\u003e\n      \u003ctd\u003e+0.5910\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd rowspan=\"6\"\u003e\u003ca href=\"https://doi.org/10.1016/j.compbiomed.2020.104121\"\u003eEffect of Deep Transfer and Multi-task Learning on Sperm Abnormality Detection\u003c/a\u003e (Nov 2020)\u003c/td\u003e\n      \u003ctd\u003eAcrosome (DTL)\u003c/td\u003e\n      \u003ctd\u003e79.00\u003c/td\u003e\n      \u003ctd\u003e80.24\u003c/td\u003e\n      \u003ctd\u003e93.42\u003c/td\u003e\n      \u003ctd\u003e82.57\u003c/td\u003e\n      \u003ctd\u003e86.58\u003c/td\u003e\n      \u003ctd\u003e79.65\u003c/td\u003e\n      \u003ctd\u003e+0.4447\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd \u003eAcrosome (DMTL)\u003c/td\u003e\n      \u003ctd\u003e80.66\u003c/td\u003e\n      \u003ctd\u003e82.42\u003c/td\u003e\n      \u003ctd\u003e92.48\u003c/td\u003e\n      \u003ctd\u003e84.26\u003c/td\u003e\n      \u003ctd\u003e87.31\u003c/td\u003e\n      \u003ctd\u003e78.19\u003c/td\u003e\n      \u003ctd\u003e+0.4984\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd \u003eHead (DTL)\u003c/td\u003e\n      \u003ctd\u003e84.00\u003c/td\u003e\n      \u003ctd\u003e87.01\u003c/td\u003e\n      \u003ctd\u003e91.78\u003c/td\u003e\n      \u003ctd\u003e87.92\u003c/td\u003e\n      \u003ctd\u003e89.36\u003c/td\u003e\n      \u003ctd\u003e81.56\u003c/td\u003e\n      \u003ctd\u003e+0.5775\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd \u003eHead (DMTL)\u003c/td\u003e\n      \u003ctd\u003e82.00\u003c/td\u003e\n      \u003ctd\u003e82.60\u003c/td\u003e\n      \u003ctd\u003e95.43\u003c/td\u003e\n      \u003ctd\u003e84.89\u003c/td\u003e\n      \u003ctd\u003e88.78\u003c/td\u003e\n      \u003ctd\u003e78.40\u003c/td\u003e\n      \u003ctd\u003e+0.5021\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd \u003eVacuole (DTL)\u003c/td\u003e\n      \u003ctd\u003e94.00\u003c/td\u003e\n      \u003ctd\u003e95.18\u003c/td\u003e\n      \u003ctd\u003e98.09\u003c/td\u003e\n      \u003ctd\u003e95.75\u003c/td\u003e\n      \u003ctd\u003e96.62\u003c/td\u003e\n      \u003ctd\u003e94.73\u003c/td\u003e\n      \u003ctd\u003e+0.7082\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eVacuole (DMTL)\u003c/td\u003e\n      \u003ctd\u003e92.33\u003c/td\u003e\n      \u003ctd\u003e94.75\u003c/td\u003e\n      \u003ctd\u003e96.56\u003c/td\u003e\n      \u003ctd\u003e95.11\u003c/td\u003e\n      \u003ctd\u003e95.65\u003c/td\u003e\n      \u003ctd\u003e93.64\u003c/td\u003e\n      \u003ctd\u003e+0.6348\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/tbody\u003e\n\u003c/table\u003e\n\n## Citation\n\nIf you use this dataset in your research, please kindly cite [our work](https://doi.org/10.1016/j.compbiomed.2019.04.030) as:\n\n```bibtex\n@article{javadi2019novel,\n  title={A novel deep learning method for automatic assessment of human sperm images},\n  author={Javadi, Soroush and Mirroshandel, Seyed Abolghasem},\n  journal={Computers in Biology and Medicine},\n  volume={109},\n  pages={182--194},\n  year={2019},\n  doi={10.1016/j.compbiomed.2019.04.030}\n}\n```\n\n## License\n\nThis dataset is made available under the [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/) license.\n\n## Credits\n\nMHSMA is based on the Human Sperm Morphology Analysis Dataset (HSMA-DS) [(Ghasemian et al., 2015)](https://doi.org/10.1016/j.cmpb.2015.08.013).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsoroushj%2Fmhsma-dataset","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsoroushj%2Fmhsma-dataset","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsoroushj%2Fmhsma-dataset/lists"}