{ "fetched_at": "2026-10-24T22:49:48Z", "source_url": "https://www.kaggle.com/competitions/bio-cas-2024-depression-detection-grand-challenge/data", "pages": [ { "name": "data-description", "content": "\n## Data Files\n* **`train.zip`** - Contains 89 clinical interview audio recordings in MP3 format.\n* **`train_json.zip`** - Provides transcriptions for the corresponding audio recordings.\n* **`test.zip`** - Contains 24 clinical interview audio recordings in MP3 format.\n* **`test_json.zip`** - Provides transcriptions for the corresponding audio recordings.\n* **train.csv** - Metadata for the training set, including audio filenames, participant gender, and age.\n* **test.csv** - Metadata for the test set, including participant IDs, gender, and age.\n* **sample_submission.csv** - A template for participants to submit their predictions in the correct format.\n\n## Columns\n\n* `Participant` - Participant ID\n* `File_name` - File name of interview recording\n* `Gender` - Gender\n* `Age` - Age\n\n## Transcripts\nTranscripts are available in JSON format, containing information such as start and end times, speaker identification, and word-level details. Each transcript entry includes the background, end, one-best transcription, speaker ID, and a list of word results.\nExample transcript entry:\n```json\n{\n \"data\": [\n {\n \"bg\": \"240\",\n \"ed\": \"1160\",\n \"onebest\": \"都去找，\",\n \"si\": \"0\",\n \"speaker\": \"1\",\n \"wordsResultList\": [\n {\n \"alternativeList\": [],\n \"wc\": \"1.0000\",\n \"wordBg\": \"4\",\n \"wordEd\": \"38\",\n \"wordsName\": \"都\",\n \"wp\": \"n\"\n },\n {\n \"alternativeList\": [],\n \"wc\": \"1.0000\",\n \"wordBg\": \"39\",\n \"wordEd\": \"55\",\n \"wordsName\": \"去\",\n \"wp\": \"n\"\n },\n {\n \"alternativeList\": [],\n \"wc\": \"1.0000\",\n \"wordBg\": \"56\",\n \"wordEd\": \"83\",\n \"wordsName\": \"找\",\n \"wp\": \"n\"\n },\n {\n \"alternativeList\": [],\n \"wc\": \"0.0000\",\n \"wordBg\": \"83\",\n \"wordEd\": \"83\",\n \"wordsName\": \"，\",\n \"wp\": \"p\"\n }\n ]\n },\n {\n \"bg\": \"1600\",\n \"ed\": \"3450\",\n \"onebest\": \"嗯没有紧张。\",\n \"si\": \"0\",\n \"speaker\": \"1\",\n \"wordsResultList\": [\n {\n \"alternativeList\": [],\n \"wc\": \"1.0000\",\n \"wordBg\": \"32\",\n \"wordEd\": \"44\",\n \"wordsName\": \"嗯\",\n \"wp\": \"s\"\n },\n {\n \"alternativeList\": [],\n \"wc\": \"1.0000\",\n \"wordBg\": \"45\",\n \"wordEd\": \"109\",\n \"wordsName\": \"没有\",\n \"wp\": \"n\"\n }\n ]\n }\n ]\n}\n```\n\nFor showcasing the parsing of JSON files, please refer to the code snippet below:\n```python\n#!/usr/bin/env python3\n# -*- coding: utf-8 -*-\n\"\"\"\nCreated on Sat May 23 21:36:02 2020\n\n@author: Kaining\n\"\"\"\nimport logging\nimport os\nimport json\nimport ast\nimport csv\nfrom contextlib import closing\n\n'''\nLoad the *.json format iFlyTek result\n'''\n\ndef load_json(file_path):\n with closing(open(file_path, \"rb\")) as file:\n data_dict = json.load(file, encoding='utf-8')\n # print(data_dict.keys())\n data_dict = ast.literal_eval(str(data_dict['data']))\n subject_id = file_path.split('/')[-1].split('.')[0]\n return data_dict, subject_id\n\n\n'''\nWrite into *.csv file\n'''\nmapping = {'1': 'Doctor', '2': 'Patient'}\n\n\ndef write_to_csv(dest_path, data_dict, subject_id):\n data_to_write = []\n tmp = []\n for key in data_dict[0].keys():\n if key == \"si\":\n continue\n tmp.append(key)\n data_to_write.append(tmp)\n\n for i in range(len(data_dict)):\n tmp = []\n for key in data_dict[i].keys():\n if key == \"si\":\n continue\n if key == \"speaker\" and data_dict[i][key] == \"1\":\n if i == 0:\n mapping['2'] = \"Doctor\"\n mapping['1'] = \"Patient\"\n tmp.append(mapping[data_dict[i][key]])\n else:\n tmp.append(mapping[data_dict[i][key]])\n elif key == \"speaker\" and data_dict[i][key] == \"2\":\n if i == 0:\n mapping['1'] = \"Doctor\"\n mapping['2'] = \"Patient\"\n tmp.append(mapping[data_dict[i][key]])\n else:\n tmp.append(mapping[data_dict[i][key]])\n elif key == \"wordsResultList\":\n tmp_2 = []\n for item in data_dict[i][\"wordsResultList\"]:\n tmp_2.append(item[\"wordsName\"].replace('\\n', '').replace('\\r', ''))\n tmp.append(tmp_2)\n else:\n tmp.append(data_dict[i][key])\n data_to_write.append(tmp)\n\n with closing(open(dest_path, 'w', encoding='utf-8')) as file:\n writer = csv.writer(file, delimiter=',')\n writer.writerows(data_to_write)\n # print(\"Finished writing\")\n logging.info(\"Finished writing\")\n\n```", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F4435490%2Fd690d196ee93da0534aa681ffd69646d%2FScreen%20Shot%202024-03-18%20at%2012.10.34%20PM.png?generation=1710785554607601&alt=media)\n\n## Data Files\n* **`train.zip`** - Contains 89 clinical interview audio recordings in MP3 format.\n* **`train_json.zip`** - Provides transcriptions for the corresponding audio recordings.\n* **`test.zip`** - Contains 24 clinical interview audio recordings in MP3 format.\n* **`test_json.zip`** - Provides transcriptions for the corresponding audio recordings.\n* **train.csv** - Metadata for the training set, including audio filenames, participant gender, and age.\n* **test.csv** - Metadata for the test set, including participant IDs, gender, and age.\n* **sample_submission.csv** - A template for participants to submit their predictions in the correct format.\n\n## Columns\n\n* `Participant` - Participant ID\n* `File_name` - File name of interview recording\n* `Gender` - Gender\n* `Age` - Age\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "## Binary Classification (Depression vs. Healthy):\nFor the binary classification task to distinguish between individuals with depression and those who are healthy, we will use the F1 score as the evaluation metric. The F1 score is calculated as:\n\n$$F_1 = 2 \\times \\frac{precision \\times recall}{precision + recall}$$\n\n​Where:\nPrecision: The proportion of true positive predictions among all positive predictions.\n$$\\text{Precision} = \\frac{TP}{TP + FP}$$\n\nRecall: The proportion of true positive predictions among all actual positives.\n$$\\text{Recall} = \\frac{TP}{TP + FN}$$\n\n## Multi-Class Classification (Depression Severity):\nFor the multi-class classification task predicting the severity of depression, we will utilize the micro-average F1 score as the evaluation metric. The micro-average F1 score calculates the average F1 score across all classes, considering the overall true positive, false positive, and false negative counts. Each class's F1 score is computed using the same formula as in binary classification.\n\n$$\n\\text{Micro-Average F1 Score} = \\frac{2 \\times {TP}}{2 \\times {TP} + {FP} + {FN}}\n$$\n\n---\n\n\n\n---\n\n## Submission File\nPlease refer to sample_submission.csv to format your results.", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Enjoy!", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "abstract", "content": "We're excited to present a unique challenge aimed at advancing automated depression diagnosis. Traditional methods using written speech or self-reported measures often fall short in real-world scenarios. To address this, we've curated a dataset of authentic depression clinical interviews from a psychiatric hospital.\n\nThe dataset includes 113 recordings (89 for training and 24 for testing), featuring interactions with 52 healthy individuals and 61 diagnosed with depression. Each participant underwent assessments using the Montgomery-Asberg Depression Rating Scale (MADRS) in Chinese, with diagnoses confirmed by psychiatry specialists.\n\nThese interviews were meticulously audio-recorded, transcribed, and annotated by experienced physicians, ensuring data quality. Participants are tasked with developing machine learning models to detect depression presence and predict severity levels using audio and text features extracted from interviews.\n\nJoin us in leveraging this groundbreaking dataset to revolutionize depression diagnosis and advance mental health care. Let's make a difference together!", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Citation", "content": "Please cite the following article in your competition submission (if applicable):\n`@article{mao2023analysis,\n author={Mao, Kaining and Wang, Deborah Baofeng and Zheng, Tiansheng and Jiao, Rongqi and Zhu, Yanhui and Wu, Bin and Qian, Lei and Lyu, Wei and Chen, Jie and Ye, Minjie},\n journal={IEEE Transactions on Biomedical Circuits and Systems}, \n title={Analysis of Automated Clinical Depression Diagnosis in a Chinese Corpus}, \n year={2023},\n volume={17},\n number={5},\n pages={1135-1152},\n keywords={Depression;Interviews;Recording;Machine learning;Multimodal sensors;Machine learning;Sentiment analysis;Medical diagnosis;Emotional corpora;machine learning;multimodal systems;nonverbal signals;sentiment analysis},\n doi={10.1109/TBCAS.2023.3291554}}`", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }