{
    "info": {
        "author": "Joseph Bullock, Carolina Cuesta-Lazaro, & Arnau Quera-Bofarull",
        "author_email": "joseph.bullock@durham.ac.uk",
        "bugtrack_url": null,
        "classifiers": [
            "License :: OSI Approved :: MIT License",
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 3"
        ],
        "description": "# XNet\n\nXNet is a Convolutional Neural Network designed for the segmentation\nof X-Ray images into bone, soft tissue and open beam\nregions. Specifically, it performs well on small datasets with the aim\nto minimise the number of false positives in the soft tissue class.\n\n## Architecture\n\n![](./Images/architecture.jpg)\n\n* Built on a typical encoder-decoder architecture as\ninspired by [SegNet](http://mi.eng.cam.ac.uk/projects/segnet/).\n\n* Additional feature extraction stage, with weight sharing across some\n  layers.\n\n* Fine and coarse grained feature preservation through concatenation\n  of layers.\n\n* L2 regularisation at each of the convolutional layers, to decrease overfitting. \n\nThe architecture is described in the ```XNet.py``` file.\n\n## Output\n\nXNet outputs a mask of equal size to the input images.\n\n![](./Images/predictions.png)\n\n## Training\n\nXNet is trained on a small dataset which has undergone\naugmention. Examples of this augmentation step can be found in the\n```augmentations.ipynb``` notebook in the ```Augmentations``` folder. Similarly the ```Training``` folder contains python scripts that perform the necessary augementations.\n\nRunning ```train.py``` from the ```Training``` folder calls various other scripts to perform one of two possible ways of augmenting the images:\n\n* 'On the fly augmentation' where a new set of augmentations is generated at each epoch.\n\n* Pre-augmented images.\n\n## Benchmarking\n\nXNet was benchmarked against two of the leading segmentation networks:\n\n* Simplified [SegNet](https://arxiv.org/abs/1511.00561) (found in the\n  ```SimpleSegNet.py``` file)\n\n* [UNet](https://arxiv.org/abs/1505.04597) (found in the ```UNet.py```\n  file)\n\n## Data\n\nWe trained on a dataset of:\n\n* 150 X-Ray images.\n\n* No scatter correction.\n\n* 1500x1500 ```.tif``` image downsampled to 200x200\n\n* 20 human body part classes.\n\n* Highly imbalanced.\n\nAs this work grew out of work with a corporation we are sadly unable to share the propriatory data we used.\n\n## More information\n\nFor more information and context see the conference poster\n```Poster.pdf```.\n\nPlease note that some of the path variables may need to be corrected in order to utilise the current filing system. These are planned to be updated in the future.\n\n",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/josephPB/XNet",
        "keywords": "",
        "license": "",
        "maintainer": "",
        "maintainer_email": "",
        "name": "durham-XNet",
        "package_url": "https://pypi.org/project/durham-XNet/",
        "platform": "",
        "project_url": "https://pypi.org/project/durham-XNet/",
        "project_urls": {
            "Homepage": "https://github.com/josephPB/XNet"
        },
        "release_url": "https://pypi.org/project/durham-XNet/0.0.4/",
        "requires_dist": [
            "durham-XNet",
            "durham-XNet.training",
            "durham-XNet.architectures",
            "durham-XNet.augmentations"
        ],
        "requires_python": "",
        "summary": "A CNN to segment X-Ray images",
        "version": "0.0.4"
    },
    "last_serial": 4807119,
    "releases": {
        "0.0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "0dce031c4f2bbbe4c254ee88bee7164a",
                    "sha256": "06bf9464e0f6232ecc9d529ed5995b00854d7d0cbb01a134fe84fc293105f30d"
                },
                "downloads": -1,
                "filename": "durham_XNet-0.0.1-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "0dce031c4f2bbbe4c254ee88bee7164a",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 36125,
                "upload_time": "2019-02-11T17:06:29",
                "url": "https://files.pythonhosted.org/packages/8c/df/be519187c6c3a0509d49b9e803fe956e6f7faca7e363073d787d977e705c/durham_XNet-0.0.1-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "3631061931718af71f67982347e8a73d",
                    "sha256": "fd33ebd275611611a4b2ba2985914450b5477ffaf185134aafd52c4286fc93d9"
                },
                "downloads": -1,
                "filename": "durham-XNet-0.0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "3631061931718af71f67982347e8a73d",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 18840,
                "upload_time": "2019-02-11T17:06:32",
                "url": "https://files.pythonhosted.org/packages/80/ce/2e7e94554fdf9c65c8de76282cbed802e6d2859fefeca70496e8bbe7be10/durham-XNet-0.0.1.tar.gz"
            }
        ],
        "0.0.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "d179a9e981074049eafd0afbc88977e1",
                    "sha256": "8711229535362caeb84b5ac2da95a226ac645ee92a70633102fc4954e97b9ab7"
                },
                "downloads": -1,
                "filename": "durham_XNet-0.0.2-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "d179a9e981074049eafd0afbc88977e1",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 57433,
                "upload_time": "2019-02-11T17:11:27",
                "url": "https://files.pythonhosted.org/packages/66/a6/d340a90d911fdf56513303d6fbe5ae5fb4accc24ea464cf72ec84fca7078/durham_XNet-0.0.2-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "793dd1c04f6d4252ce8609f144443bb7",
                    "sha256": "44d79cc9109f2b79e6cd02a81c7df50d6b4a7c470dd3368349d11f562319ccbf"
                },
                "downloads": -1,
                "filename": "durham-XNet-0.0.2.tar.gz",
                "has_sig": false,
                "md5_digest": "793dd1c04f6d4252ce8609f144443bb7",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 18845,
                "upload_time": "2019-02-11T17:11:29",
                "url": "https://files.pythonhosted.org/packages/f8/50/827d6be3ddd141de81cadcd67df8250cc14e7331765f68350e056455835b/durham-XNet-0.0.2.tar.gz"
            }
        ],
        "0.0.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "b5946dd9db425d331e8a69428a5caf66",
                    "sha256": "65ed876d5801202aaf788d57c398c06e5c067e7476db2ff7710361b43d421d03"
                },
                "downloads": -1,
                "filename": "durham_XNet-0.0.3-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "b5946dd9db425d331e8a69428a5caf66",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 36381,
                "upload_time": "2019-02-11T17:17:38",
                "url": "https://files.pythonhosted.org/packages/87/cd/1d4fbe3a45611feaf7eabaae71d91b8ede5e1755f7c8eb4b638bf8f0b0ed/durham_XNet-0.0.3-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "e55046e72de97be607d45cf2a0b1ddfa",
                    "sha256": "8fa3aea589816a584fe9e1ed5792816f4645dd1349f91efc8788b9d0287ccf98"
                },
                "downloads": -1,
                "filename": "durham-XNet-0.0.3.tar.gz",
                "has_sig": false,
                "md5_digest": "e55046e72de97be607d45cf2a0b1ddfa",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 18854,
                "upload_time": "2019-02-11T17:17:40",
                "url": "https://files.pythonhosted.org/packages/11/db/9b5ac3071ca3a9832fe0b1be6d82268c4cf0500b01bbba37512b67fc607d/durham-XNet-0.0.3.tar.gz"
            }
        ],
        "0.0.4": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "f4ac916617d853dbc23e5e24e4c56f20",
                    "sha256": "017d1bdba09cf99c97f6caf584dc1d621560a6091227f2cee25708b5dc579319"
                },
                "downloads": -1,
                "filename": "durham_XNet-0.0.4-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "f4ac916617d853dbc23e5e24e4c56f20",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 36446,
                "upload_time": "2019-02-11T17:20:31",
                "url": "https://files.pythonhosted.org/packages/87/24/81a907f0434011030f1dbcb23c329fbb3a1f32f62c227e40a1141bb11f65/durham_XNet-0.0.4-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "e1cf83450d909dd3579c9e8a62b883f8",
                    "sha256": "770d952dc0302fdb890ba6a7f58eeca3db94cc96c78594f2006b59dba2ffe8dc"
                },
                "downloads": -1,
                "filename": "durham-XNet-0.0.4.tar.gz",
                "has_sig": false,
                "md5_digest": "e1cf83450d909dd3579c9e8a62b883f8",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 18855,
                "upload_time": "2019-02-11T17:20:32",
                "url": "https://files.pythonhosted.org/packages/76/49/d0e07ca3b4782a75ebb3ee34c318b883d091b2e74fbf676fca7735252b3b/durham-XNet-0.0.4.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "f4ac916617d853dbc23e5e24e4c56f20",
                "sha256": "017d1bdba09cf99c97f6caf584dc1d621560a6091227f2cee25708b5dc579319"
            },
            "downloads": -1,
            "filename": "durham_XNet-0.0.4-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "f4ac916617d853dbc23e5e24e4c56f20",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 36446,
            "upload_time": "2019-02-11T17:20:31",
            "url": "https://files.pythonhosted.org/packages/87/24/81a907f0434011030f1dbcb23c329fbb3a1f32f62c227e40a1141bb11f65/durham_XNet-0.0.4-py3-none-any.whl"
        },
        {
            "comment_text": "",
            "digests": {
                "md5": "e1cf83450d909dd3579c9e8a62b883f8",
                "sha256": "770d952dc0302fdb890ba6a7f58eeca3db94cc96c78594f2006b59dba2ffe8dc"
            },
            "downloads": -1,
            "filename": "durham-XNet-0.0.4.tar.gz",
            "has_sig": false,
            "md5_digest": "e1cf83450d909dd3579c9e8a62b883f8",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 18855,
            "upload_time": "2019-02-11T17:20:32",
            "url": "https://files.pythonhosted.org/packages/76/49/d0e07ca3b4782a75ebb3ee34c318b883d091b2e74fbf676fca7735252b3b/durham-XNet-0.0.4.tar.gz"
        }
    ]
}