From 1e09865d7ea3ff41d38ac2794924055e0962468d Mon Sep 17 00:00:00 2001 From: William Yu Jackson Date: Sat, 20 Sep 2025 14:16:55 +1000 Subject: [PATCH] Added synthetic data schemas, sample datasets, preprocessing notebook and readme --- AI Guardian/DataSchemas_WJ/README.MD | 47 +++ .../DataSchemas_WJ/preprocessing.ipynb | 288 ++++++++++++++++++ .../emotions_clean-checkpoint.csv | 11 + .../processed_data/emotions_clean.csv | 11 + .../processed_data/merged_timeseries.csv | 21 ++ .../processed_data/vitals_clean.csv | 21 ++ .../processed_data/wearables_clean.csv | 21 ++ .../sample_vital-checkpoint.csv | 21 ++ .../sample_wearable-checkpoint.csv | 21 ++ .../schema_alert-checkpoint.csv | 10 + .../schema_vital-checkpoint.csv | 11 + .../schema_wearable-checkpoint.csv | 11 + .../raw_data/sample_emotion.csv | 11 + .../DataSchemas_WJ/raw_data/sample_vital.csv | 21 ++ .../raw_data/sample_wearable.csv | 21 ++ .../DataSchemas_WJ/raw_data/schema_alert.csv | 10 + .../DataSchemas_WJ/raw_data/schema_vital.csv | 11 + .../raw_data/schema_wearable.csv | 11 + 18 files changed, 579 insertions(+) create mode 100644 AI Guardian/DataSchemas_WJ/README.MD create mode 100644 AI Guardian/DataSchemas_WJ/preprocessing.ipynb create mode 100644 AI Guardian/DataSchemas_WJ/processed_data/.ipynb_checkpoints/emotions_clean-checkpoint.csv create mode 100644 AI Guardian/DataSchemas_WJ/processed_data/emotions_clean.csv create mode 100644 AI Guardian/DataSchemas_WJ/processed_data/merged_timeseries.csv create mode 100644 AI Guardian/DataSchemas_WJ/processed_data/vitals_clean.csv create mode 100644 AI Guardian/DataSchemas_WJ/processed_data/wearables_clean.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_vital-checkpoint.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_wearable-checkpoint.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_alert-checkpoint.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_vital-checkpoint.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_wearable-checkpoint.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/sample_emotion.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/sample_vital.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/sample_wearable.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/schema_alert.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/schema_vital.csv create mode 100644 AI Guardian/DataSchemas_WJ/raw_data/schema_wearable.csv diff --git a/AI Guardian/DataSchemas_WJ/README.MD b/AI Guardian/DataSchemas_WJ/README.MD new file mode 100644 index 00000000..06ac3116 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/README.MD @@ -0,0 +1,47 @@ +# Data Schemas & Sample Datasets – William Jackson + +## Overview + +This folder is for the Guardian alert system with emotional and healht risk detection - data & Requirement\_ task. +It contains reference schemas, sample datasets, and preprocessing notebooks for vitals, wearable inputs, and optional emotion data. + +## Purpose + +- Define data schemas for vitals, wearable inputs, and optional emotion data (facial/voice). +- Collect annotated vitals time-series and simulated wearable datasets. +- Design preprocessing pipelines for time-series cleaning and baseline normalization. +- Establish privacy and security guidelines for wearable and sensitive emotion data. +- Develop data acquisition plan including integration of future wearable data sources. +- Implement quality assurance for time-series integrity and labeling accuracy. +- Review requirements with clinical staff and guardians for alert relevance. + +## Contents + +- **Sample datasets** + + - `sample_emotion.csv` – example of optional emotion input (facial/voice tags). + - `sample_vital.csv` – simulated vital signs time-series. + - `sample_wearable.csv` – wearable activity and lifestyle signals. + +- **Schemas** + + - `schema_alert.csv` – structure of system-generated alerts. + - `schema_vital.csv` – expected fields for vitals ingestion. + - `schema_wearable.csv` – expected fields for wearable inputs. + +- **Notebooks** + - `preprocessing.ipynb` – data cleaning, normalization, schema validation. + +--- + +## How to Use + +1. Review `schemas/` to understand the required fields. +2. Load files from `raw_data/` as test inputs. +3. Use `preprocessing.ipynb` to validate and normalize sample data. +4. Save cleaned outputs into `processed_data/`. + +--- + +_Author: William Jackson_ +_AI Sprint T2 2025 – Data & Requirements_ diff --git a/AI Guardian/DataSchemas_WJ/preprocessing.ipynb b/AI Guardian/DataSchemas_WJ/preprocessing.ipynb new file mode 100644 index 00000000..07dd4ef4 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/preprocessing.ipynb @@ -0,0 +1,288 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "97f464ab-7166-4a19-83fa-3b016d563a50", + "metadata": {}, + "source": [ + "# PreProcess and sanity checks" + ] + }, + { + "cell_type": "markdown", + "id": "eef78850-72c1-4620-90ec-3623b2700ed7", + "metadata": {}, + "source": [ + "Import lib" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "c7396eae-6f9e-4f5c-8313-0b5dd6d49e24", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "import pandas as pd\n", + "import matplotlib.pyplot as plt" + ] + }, + { + "cell_type": "markdown", + "id": "2f407eec-8485-436c-9509-4a5ba7d88be6", + "metadata": {}, + "source": [ + "Safety check" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "5c8c68d6-6dbc-47e4-bd0f-64f0fde63038", + "metadata": {}, + "outputs": [], + "source": [ + "DATA = \"data\"\n", + "if not os.path.exists(DATA):\n", + " raise FileNotFoundError(f\"Expected data folder at {DATA}, but it was not found.\")" + ] + }, + { + "cell_type": "markdown", + "id": "ba0e6235-e2d5-44c3-864b-6de83d8d7d6a", + "metadata": {}, + "source": [ + "load the csvs" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "5736b04a-586c-4e1b-9fb1-6cc638c6b527", + "metadata": {}, + "outputs": [], + "source": [ + "vitals = pd.read_csv(f\"{DATA}/sample_vital.csv\", parse_dates=[\"timestamp\"])\n", + "wear = pd.read_csv(f\"{DATA}/sample_wearable.csv\", parse_dates=[\"timestamp\"])\n", + "emo = pd.read_csv(f\"{DATA}/sample_emotion.csv\", parse_dates=[\"timestamp\"])\n" + ] + }, + { + "cell_type": "markdown", + "id": "a8190491-ecf5-4aab-b4b3-91508947a73f", + "metadata": {}, + "source": [ + "print the columns" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "c7dcaa91-8eb7-4c5e-8c59-1ee7d24953c2", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "vitals cols: ['patient_id', 'timestamp', 'heart_rate', 'spo2', 'resp_rate', 'temp', 'systolic_bp', 'diastolic_bp', 'device', 'quality_flag']\n", + "wear cols: ['patient_id', 'timestamp', 'step_count', 'activity_level', 'accel_x', 'accel_y', 'accel_z', 'calories', 'device', 'quality_flag']\n", + "emo cols: ['patient_id', 'timestamp', 'source', 'emotion', 'confidence', 'notes', 'model', 'privacy_tag']\n", + "rows: 20 20 10\n" + ] + } + ], + "source": [ + "print(\"vitals cols:\", list(vitals.columns))\n", + "print(\"wear cols:\", list(wear.columns))\n", + "print(\"emo cols:\", list(emo.columns))\n", + "print(\"rows:\", len(vitals), len(wear), len(emo))" + ] + }, + { + "cell_type": "markdown", + "id": "df00ea50-44ea-4491-ace0-3e01e8e5c855", + "metadata": {}, + "source": [ + "Filter to enforece data quality based on schema." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "a3c36dff-5c85-40ff-b596-9fe99db63d34", + "metadata": {}, + "outputs": [], + "source": [ + "vitals = vitals.query(\n", + " \"30 <= heart_rate <= 220 and 70 <= spo2 <= 100 and 4 <= resp_rate <= 60 and 30 <= temp <= 43\"\n", + ")\n", + "wear = wear.query(\n", + " \"0 <= step_count <= 50000 and 0 <= calories <= 10000 and activity_level in ['sedentary','light','moderate','intense']\"\n", + ")\n", + "emo = emo.query(\n", + " \"0.0 <= confidence <= 1.0 and source in ['face','voice','self_report'] and privacy_tag in ['normal','sensitive','restricted']\"\n", + ")\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "id": "fb6e7810-c949-47c0-b480-7150036cd69b", + "metadata": {}, + "source": [ + "scales the varaibles to a 1 to 0 range which would make it simpler to compare signals on a chart. For example steps vs bpm would be on complete different scales" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "3a2bd96e-966a-4dc5-af31-b17e9f1eb881", + "metadata": {}, + "outputs": [], + "source": [ + "vitals[\"hr_norm\"] = (vitals[\"heart_rate\"] - 30) / (220 - 30)\n", + "vitals[\"spo2_norm\"] = (vitals[\"spo2\"] - 70) / (100 - 70)\n", + "wear[\"steps_norm\"] = wear[\"step_count\"] / 50000\n", + "wear[\"cal_norm\"] = wear[\"calories\"] / 10000\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "id": "e12fb126-ca6a-487c-a73b-b010b21caa8d", + "metadata": {}, + "source": [ + "aligns wearable data and vitals by patient and timestamp to make sure only rows where both srouces exist" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "3be11f6c-3f11-4aa8-ad5f-04a131ce2115", + "metadata": {}, + "outputs": [], + "source": [ + "merged = pd.merge(\n", + " vitals, wear,\n", + " on=[\"patient_id\", \"timestamp\"],\n", + " how=\"inner\",\n", + " suffixes=(\"_v\", \"_w\")\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "649a95e6-4a7b-4139-b9c2-be80026d3e68", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "written and cleaned/\n" + ] + } + ], + "source": [ + "os.makedirs(\"cleaned\", exist_ok=True)\n", + "vitals.to_csv(\"cleaned/vitals_clean.csv\", index=False)\n", + "wear.to_csv(\"cleaned/wearables_clean.csv\", index=False)\n", + "emo.to_csv(\"cleaned/emotions_clean.csv\", index=False)\n", + "merged.to_csv(\"cleaned/merged_timeseries.csv\", index=False)\n", + "print(\"written and cleaned/\")" + ] + }, + { + "cell_type": "markdown", + "id": "d99d80a8-4408-4432-a3d2-a0a75ca3d4b5", + "metadata": {}, + "source": [ + "Here we create a plot for a easy visual analysis to see wheter the cleaned vital data looks normal" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "da17afce-d3e3-4530-8993-df30cd33f8e3", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "for pid, chunk in vitals.groupby(\"patient_id\"):\n", + " plt.figure()\n", + " plt.plot(chunk[\"timestamp\"], chunk[\"heart_rate\"])\n", + " plt.title(f\"heart rate – {pid}\")\n", + " plt.xlabel(\"time\"); plt.ylabel(\"bpm\"); plt.xticks(rotation=30)\n", + " plt.tight_layout(); plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "47200331-d9e0-4302-8bc9-f83e19738f6a", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "668844b8-a8bf-4d2a-8586-ee9517f2a308", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c117d538-0ed9-4f2b-af65-3d9e4a0a9b47", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python (base)", + "language": "python", + "name": "base" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.7" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/AI Guardian/DataSchemas_WJ/processed_data/.ipynb_checkpoints/emotions_clean-checkpoint.csv b/AI Guardian/DataSchemas_WJ/processed_data/.ipynb_checkpoints/emotions_clean-checkpoint.csv new file mode 100644 index 00000000..c3a19692 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/processed_data/.ipynb_checkpoints/emotions_clean-checkpoint.csv @@ -0,0 +1,11 @@ +patient_id,timestamp,source,emotion,confidence,notes,model,privacy_tag +P001,2000-01-01 09:00:00+00:00,face,neutral,0.91,,emoNet-v2,normal +P001,2000-01-01 09:05:00+00:00,voice,stressed,0.77,,emoNet-v2,normal +P001,2000-01-01 09:10:00+00:00,face,happy,0.85,,emoNet-v2,normal +P001,2000-01-01 09:15:00+00:00,voice,tired,0.66,,emoNet-v2,normal +P001,2000-01-01 09:20:00+00:00,face,angry,0.72,,emoNet-v2,normal +P002,2000-01-01 09:25:00+00:00,voice,neutral,0.88,,emoNet-v2,normal +P002,2000-01-01 09:30:00+00:00,face,surprise,0.79,,emoNet-v2,normal +P002,2000-01-01 09:35:00+00:00,voice,sad,0.81,,emoNet-v2,normal +P002,2000-01-01 09:40:00+00:00,face,happy,0.92,,emoNet-v2,normal +P002,2000-01-01 09:45:00+00:00,voice,stressed,0.84,,emoNet-v2,normal diff --git a/AI Guardian/DataSchemas_WJ/processed_data/emotions_clean.csv b/AI Guardian/DataSchemas_WJ/processed_data/emotions_clean.csv new file mode 100644 index 00000000..bbf88a0e --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/processed_data/emotions_clean.csv @@ -0,0 +1,11 @@ +patient_id,timestamp,source,emotion,confidence,notes,model,privacy_tag +P001,2000-01-01 09:00:00+00:00,face,neutral,0.91,,v1,normal +P001,2000-01-01 09:05:00+00:00,voice,stressed,0.77,,v1,normal +P001,2000-01-01 09:10:00+00:00,face,happy,0.85,,v1,normal +P001,2000-01-01 09:15:00+00:00,voice,tired,0.66,,v1,normal +P001,2000-01-01 09:20:00+00:00,face,angry,0.72,,v1,normal +P002,2000-01-01 09:25:00+00:00,voice,neutral,0.88,,v1,normal +P002,2000-01-01 09:30:00+00:00,face,surprise,0.79,,v1,normal +P002,2000-01-01 09:35:00+00:00,voice,sad,0.81,,v1,normal +P002,2000-01-01 09:40:00+00:00,face,happy,0.92,,v1,normal +P002,2000-01-01 09:45:00+00:00,voice,stressed,0.84,,v1,normal diff --git a/AI Guardian/DataSchemas_WJ/processed_data/merged_timeseries.csv b/AI Guardian/DataSchemas_WJ/processed_data/merged_timeseries.csv new file mode 100644 index 00000000..e81c6dfc --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/processed_data/merged_timeseries.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,heart_rate,spo2,resp_rate,temp,systolic_bp,diastolic_bp,device_v,quality_flag_v,hr_norm,spo2_norm,step_count,activity_level,accel_x,accel_y,accel_z,calories,device_w,quality_flag_w,steps_norm,cal_norm +P001,2000-01-01 09:00:00+00:00,86,96,17,36.8,125,82,Fitbit,ok,0.29473684210526313,0.8666666666666667,0,sedentary,0.05,-0.01,9.77,1.5,Fitbit,ok,0.0,0.00015 +P001,2000-01-01 09:05:00+00:00,91,98,15,36.6,117,85,Fitbit,ok,0.32105263157894737,0.9333333333333333,45,light,0.1,0.03,9.85,2.3,Fitbit,ok,0.0009,0.00022999999999999998 +P001,2000-01-01 09:10:00+00:00,77,95,18,37.0,138,73,Fitbit,ok,0.24736842105263157,0.8333333333333334,120,moderate,0.08,0.02,9.7,3.8,Fitbit,ok,0.0024,0.00037999999999999997 +P001,2000-01-01 09:15:00+00:00,101,99,16,36.9,133,78,Fitbit,ok,0.3736842105263158,0.9666666666666667,230,intense,-0.12,0.04,9.92,5.6,Fitbit,ok,0.0046,0.00056 +P001,2000-01-01 09:20:00+00:00,84,97,14,36.5,121,87,Fitbit,ok,0.28421052631578947,0.9,60,light,0.02,-0.05,9.81,2.7,Fitbit,ok,0.0012,0.00027 +P001,2000-01-01 09:25:00+00:00,92,96,19,37.2,130,74,Fitbit,ok,0.3263157894736842,0.8666666666666667,180,moderate,0.06,0.01,9.76,4.9,Fitbit,ok,0.0036,0.00049 +P001,2000-01-01 09:30:00+00:00,79,98,13,36.7,116,81,Fitbit,ok,0.2578947368421053,0.9333333333333333,15,sedentary,-0.04,0.02,9.79,1.8,Fitbit,ok,0.0003,0.00018 +P001,2000-01-01 09:35:00+00:00,88,94,15,36.9,127,75,Fitbit,ok,0.30526315789473685,0.8,95,light,0.09,-0.03,9.88,3.2,Fitbit,ok,0.0019,0.00032 +P001,2000-01-01 09:40:00+00:00,95,97,17,36.6,119,84,Fitbit,ok,0.34210526315789475,0.9,310,intense,0.14,0.05,9.83,6.4,Fitbit,ok,0.0062,0.00064 +P001,2000-01-01 09:45:00+00:00,83,99,16,36.8,122,80,Fitbit,ok,0.2789473684210526,0.9666666666666667,150,moderate,-0.08,0.06,9.74,4.2,Fitbit,ok,0.003,0.00042 +P002,2000-01-01 09:00:00+00:00,74,97,16,36.7,121,78,Apple Watch,ok,0.23157894736842105,0.9,20,light,0.04,0.01,9.79,2.0,Apple Watch,ok,0.0004,0.0002 +P002,2000-01-01 09:05:00+00:00,88,96,14,36.8,118,82,Apple Watch,ok,0.30526315789473685,0.8666666666666667,80,moderate,-0.06,0.02,9.82,3.1,Apple Watch,ok,0.0016,0.00031 +P002,2000-01-01 09:10:00+00:00,81,95,17,36.6,125,85,Apple Watch,ok,0.26842105263157895,0.8333333333333334,200,intense,0.11,-0.04,9.75,6.0,Apple Watch,ok,0.004,0.0006 +P002,2000-01-01 09:15:00+00:00,90,98,15,36.9,119,76,Apple Watch,ok,0.3157894736842105,0.9333333333333333,35,light,0.02,0.0,9.8,2.4,Apple Watch,ok,0.0007,0.00023999999999999998 +P002,2000-01-01 09:20:00+00:00,85,99,18,36.7,123,80,Apple Watch,ok,0.2894736842105263,0.9666666666666667,145,moderate,-0.05,0.03,9.84,4.7,Apple Watch,ok,0.0029,0.00047000000000000004 +P002,2000-01-01 09:25:00+00:00,93,96,16,36.8,127,83,Apple Watch,ok,0.33157894736842103,0.8666666666666667,250,intense,0.13,-0.02,9.77,7.2,Apple Watch,ok,0.005,0.00072 +P002,2000-01-01 09:30:00+00:00,76,94,14,36.5,122,79,Apple Watch,ok,0.24210526315789474,0.8,10,sedentary,0.0,0.01,9.81,1.6,Apple Watch,ok,0.0002,0.00016 +P002,2000-01-01 09:35:00+00:00,89,97,17,36.9,124,81,Apple Watch,ok,0.3105263157894737,0.9,95,light,0.07,-0.01,9.85,3.3,Apple Watch,ok,0.0019,0.00033 +P002,2000-01-01 09:40:00+00:00,82,98,15,36.7,120,77,Apple Watch,ok,0.2736842105263158,0.9333333333333333,310,intense,-0.09,0.04,9.78,6.5,Apple Watch,ok,0.0062,0.00065 +P002,2000-01-01 09:45:00+00:00,87,95,16,36.6,126,84,Apple Watch,ok,0.3,0.8333333333333334,175,moderate,0.05,0.06,9.82,4.8,Apple Watch,ok,0.0035,0.00047999999999999996 diff --git a/AI Guardian/DataSchemas_WJ/processed_data/vitals_clean.csv b/AI Guardian/DataSchemas_WJ/processed_data/vitals_clean.csv new file mode 100644 index 00000000..d8329b6f --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/processed_data/vitals_clean.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,heart_rate,spo2,resp_rate,temp,systolic_bp,diastolic_bp,device,quality_flag,hr_norm,spo2_norm +P001,2000-01-01 09:00:00+00:00,86,96,17,36.8,125,82,Fitbit,ok,0.29473684210526313,0.8666666666666667 +P001,2000-01-01 09:05:00+00:00,91,98,15,36.6,117,85,Fitbit,ok,0.32105263157894737,0.9333333333333333 +P001,2000-01-01 09:10:00+00:00,77,95,18,37.0,138,73,Fitbit,ok,0.24736842105263157,0.8333333333333334 +P001,2000-01-01 09:15:00+00:00,101,99,16,36.9,133,78,Fitbit,ok,0.3736842105263158,0.9666666666666667 +P001,2000-01-01 09:20:00+00:00,84,97,14,36.5,121,87,Fitbit,ok,0.28421052631578947,0.9 +P001,2000-01-01 09:25:00+00:00,92,96,19,37.2,130,74,Fitbit,ok,0.3263157894736842,0.8666666666666667 +P001,2000-01-01 09:30:00+00:00,79,98,13,36.7,116,81,Fitbit,ok,0.2578947368421053,0.9333333333333333 +P001,2000-01-01 09:35:00+00:00,88,94,15,36.9,127,75,Fitbit,ok,0.30526315789473685,0.8 +P001,2000-01-01 09:40:00+00:00,95,97,17,36.6,119,84,Fitbit,ok,0.34210526315789475,0.9 +P001,2000-01-01 09:45:00+00:00,83,99,16,36.8,122,80,Fitbit,ok,0.2789473684210526,0.9666666666666667 +P002,2000-01-01 09:00:00+00:00,74,97,16,36.7,121,78,Apple Watch,ok,0.23157894736842105,0.9 +P002,2000-01-01 09:05:00+00:00,88,96,14,36.8,118,82,Apple Watch,ok,0.30526315789473685,0.8666666666666667 +P002,2000-01-01 09:10:00+00:00,81,95,17,36.6,125,85,Apple Watch,ok,0.26842105263157895,0.8333333333333334 +P002,2000-01-01 09:15:00+00:00,90,98,15,36.9,119,76,Apple Watch,ok,0.3157894736842105,0.9333333333333333 +P002,2000-01-01 09:20:00+00:00,85,99,18,36.7,123,80,Apple Watch,ok,0.2894736842105263,0.9666666666666667 +P002,2000-01-01 09:25:00+00:00,93,96,16,36.8,127,83,Apple Watch,ok,0.33157894736842103,0.8666666666666667 +P002,2000-01-01 09:30:00+00:00,76,94,14,36.5,122,79,Apple Watch,ok,0.24210526315789474,0.8 +P002,2000-01-01 09:35:00+00:00,89,97,17,36.9,124,81,Apple Watch,ok,0.3105263157894737,0.9 +P002,2000-01-01 09:40:00+00:00,82,98,15,36.7,120,77,Apple Watch,ok,0.2736842105263158,0.9333333333333333 +P002,2000-01-01 09:45:00+00:00,87,95,16,36.6,126,84,Apple Watch,ok,0.3,0.8333333333333334 diff --git a/AI Guardian/DataSchemas_WJ/processed_data/wearables_clean.csv b/AI Guardian/DataSchemas_WJ/processed_data/wearables_clean.csv new file mode 100644 index 00000000..c10f50df --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/processed_data/wearables_clean.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,step_count,activity_level,accel_x,accel_y,accel_z,calories,device,quality_flag,steps_norm,cal_norm +P001,2000-01-01 09:00:00+00:00,0,sedentary,0.05,-0.01,9.77,1.5,Fitbit,ok,0.0,0.00015 +P001,2000-01-01 09:05:00+00:00,45,light,0.1,0.03,9.85,2.3,Fitbit,ok,0.0009,0.00022999999999999998 +P001,2000-01-01 09:10:00+00:00,120,moderate,0.08,0.02,9.7,3.8,Fitbit,ok,0.0024,0.00037999999999999997 +P001,2000-01-01 09:15:00+00:00,230,intense,-0.12,0.04,9.92,5.6,Fitbit,ok,0.0046,0.00056 +P001,2000-01-01 09:20:00+00:00,60,light,0.02,-0.05,9.81,2.7,Fitbit,ok,0.0012,0.00027 +P001,2000-01-01 09:25:00+00:00,180,moderate,0.06,0.01,9.76,4.9,Fitbit,ok,0.0036,0.00049 +P001,2000-01-01 09:30:00+00:00,15,sedentary,-0.04,0.02,9.79,1.8,Fitbit,ok,0.0003,0.00018 +P001,2000-01-01 09:35:00+00:00,95,light,0.09,-0.03,9.88,3.2,Fitbit,ok,0.0019,0.00032 +P001,2000-01-01 09:40:00+00:00,310,intense,0.14,0.05,9.83,6.4,Fitbit,ok,0.0062,0.00064 +P001,2000-01-01 09:45:00+00:00,150,moderate,-0.08,0.06,9.74,4.2,Fitbit,ok,0.003,0.00042 +P002,2000-01-01 09:00:00+00:00,20,light,0.04,0.01,9.79,2.0,Apple Watch,ok,0.0004,0.0002 +P002,2000-01-01 09:05:00+00:00,80,moderate,-0.06,0.02,9.82,3.1,Apple Watch,ok,0.0016,0.00031 +P002,2000-01-01 09:10:00+00:00,200,intense,0.11,-0.04,9.75,6.0,Apple Watch,ok,0.004,0.0006 +P002,2000-01-01 09:15:00+00:00,35,light,0.02,0.0,9.8,2.4,Apple Watch,ok,0.0007,0.00023999999999999998 +P002,2000-01-01 09:20:00+00:00,145,moderate,-0.05,0.03,9.84,4.7,Apple Watch,ok,0.0029,0.00047000000000000004 +P002,2000-01-01 09:25:00+00:00,250,intense,0.13,-0.02,9.77,7.2,Apple Watch,ok,0.005,0.00072 +P002,2000-01-01 09:30:00+00:00,10,sedentary,0.0,0.01,9.81,1.6,Apple Watch,ok,0.0002,0.00016 +P002,2000-01-01 09:35:00+00:00,95,light,0.07,-0.01,9.85,3.3,Apple Watch,ok,0.0019,0.00033 +P002,2000-01-01 09:40:00+00:00,310,intense,-0.09,0.04,9.78,6.5,Apple Watch,ok,0.0062,0.00065 +P002,2000-01-01 09:45:00+00:00,175,moderate,0.05,0.06,9.82,4.8,Apple Watch,ok,0.0035,0.00047999999999999996 diff --git a/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_vital-checkpoint.csv b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_vital-checkpoint.csv new file mode 100644 index 00000000..f61f0507 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_vital-checkpoint.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,heart_rate,spo2,resp_rate,temp,systolic_bp,diastolic_bp,device,quality_flag +P001,2000-01-01T09:00:00Z,86,96,17,36.8,125,82,Fitbit,ok +P001,2000-01-01T09:05:00Z,91,98,15,36.6,117,85,Fitbit,ok +P001,2000-01-01T09:10:00Z,77,95,18,37.0,138,73,Fitbit,ok +P001,2000-01-01T09:15:00Z,101,99,16,36.9,133,78,Fitbit,ok +P001,2000-01-01T09:20:00Z,84,97,14,36.5,121,87,Fitbit,ok +P001,2000-01-01T09:25:00Z,92,96,19,37.2,130,74,Fitbit,ok +P001,2000-01-01T09:30:00Z,79,98,13,36.7,116,81,Fitbit,ok +P001,2000-01-01T09:35:00Z,88,94,15,36.9,127,75,Fitbit,ok +P001,2000-01-01T09:40:00Z,95,97,17,36.6,119,84,Fitbit,ok +P001,2000-01-01T09:45:00Z,83,99,16,36.8,122,80,Fitbit,ok +P002,2000-01-01T09:00:00Z,74,97,16,36.7,121,78,Apple Watch,ok +P002,2000-01-01T09:05:00Z,88,96,14,36.8,118,82,Apple Watch,ok +P002,2000-01-01T09:10:00Z,81,95,17,36.6,125,85,Apple Watch,ok +P002,2000-01-01T09:15:00Z,90,98,15,36.9,119,76,Apple Watch,ok +P002,2000-01-01T09:20:00Z,85,99,18,36.7,123,80,Apple Watch,ok +P002,2000-01-01T09:25:00Z,93,96,16,36.8,127,83,Apple Watch,ok +P002,2000-01-01T09:30:00Z,76,94,14,36.5,122,79,Apple Watch,ok +P002,2000-01-01T09:35:00Z,89,97,17,36.9,124,81,Apple Watch,ok +P002,2000-01-01T09:40:00Z,82,98,15,36.7,120,77,Apple Watch,ok +P002,2000-01-01T09:45:00Z,87,95,16,36.6,126,84,Apple Watch,ok diff --git a/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_wearable-checkpoint.csv b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_wearable-checkpoint.csv new file mode 100644 index 00000000..1b3822bb --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/sample_wearable-checkpoint.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,step_count,activity_level,accel_x,accel_y,accel_z,calories,device,quality_flag +P001,2000-01-01T09:00:00Z,0,sedentary,0.05,-0.01,9.77,1.5,Fitbit,ok +P001,2000-01-01T09:05:00Z,45,light,0.10,0.03,9.85,2.3,Fitbit,ok +P001,2000-01-01T09:10:00Z,120,moderate,0.08,0.02,9.70,3.8,Fitbit,ok +P001,2000-01-01T09:15:00Z,230,intense,-0.12,0.04,9.92,5.6,Fitbit,ok +P001,2000-01-01T09:20:00Z,60,light,0.02,-0.05,9.81,2.7,Fitbit,ok +P001,2000-01-01T09:25:00Z,180,moderate,0.06,0.01,9.76,4.9,Fitbit,ok +P001,2000-01-01T09:30:00Z,15,sedentary,-0.04,0.02,9.79,1.8,Fitbit,ok +P001,2000-01-01T09:35:00Z,95,light,0.09,-0.03,9.88,3.2,Fitbit,ok +P001,2000-01-01T09:40:00Z,310,intense,0.14,0.05,9.83,6.4,Fitbit,ok +P001,2000-01-01T09:45:00Z,150,moderate,-0.08,0.06,9.74,4.2,Fitbit,ok +P002,2000-01-01T09:00:00Z,20,light,0.04,0.01,9.79,2.0,Apple Watch,ok +P002,2000-01-01T09:05:00Z,80,moderate,-0.06,0.02,9.82,3.1,Apple Watch,ok +P002,2000-01-01T09:10:00Z,200,intense,0.11,-0.04,9.75,6.0,Apple Watch,ok +P002,2000-01-01T09:15:00Z,35,light,0.02,0.00,9.80,2.4,Apple Watch,ok +P002,2000-01-01T09:20:00Z,145,moderate,-0.05,0.03,9.84,4.7,Apple Watch,ok +P002,2000-01-01T09:25:00Z,250,intense,0.13,-0.02,9.77,7.2,Apple Watch,ok +P002,2000-01-01T09:30:00Z,10,sedentary,0.00,0.01,9.81,1.6,Apple Watch,ok +P002,2000-01-01T09:35:00Z,95,light,0.07,-0.01,9.85,3.3,Apple Watch,ok +P002,2000-01-01T09:40:00Z,310,intense,-0.09,0.04,9.78,6.5,Apple Watch,ok +P002,2000-01-01T09:45:00Z,175,moderate,0.05,0.06,9.82,4.8,Apple Watch,ok diff --git a/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_alert-checkpoint.csv b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_alert-checkpoint.csv new file mode 100644 index 00000000..7a13ddb6 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_alert-checkpoint.csv @@ -0,0 +1,10 @@ +field_name,type,required,description,example,range +alert_id,string,yes,Unique alert ID,A-0001, +patient_id,string,yes,Patient this alert is about,P001, +timestamp,datetime,yes,Alert creation time,2025-08-22T10:00:00Z, +alert_type,string,yes,Type of alert,tachycardia,tachycardia|hypoxia|fever|bradycardia|hypertension|hypotension|emotion_risk +severity,string,yes,How serious it is,medium,low|medium|high|critical +trigger_fields,string,no,Fields that triggered alert,"heart_rate,spo2", +rule_id,string,no,Rule that caused alert,R-HEART-01, +ack_status,string,no,If alert has been acknowledged,unread,unread|acknowledged|resolved +notes,string,no,Free text for reviewer,Short run post-exercise, diff --git a/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_vital-checkpoint.csv b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_vital-checkpoint.csv new file mode 100644 index 00000000..30e1ed53 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_vital-checkpoint.csv @@ -0,0 +1,11 @@ +field_name,type,required,description,example,range +patient_id,string,yes,Unique ID for the patient,P001, +timestamp,datetime,yes,Time recorded in UTC,2025-08-22T09:30:00Z, +heart_rate,int,yes,Beats per minute,72,30-220 +spo2,int,yes,Oxygen % level,98,70-100 +resp_rate,int,no,Breaths per min,16,4-60 +temp,float,no,Body temp in Celsius,36.8,30-43 +systolic_bp,int,no,Top blood pressure value,118,60-260 +diastolic_bp,int,no,Bottom blood pressure value,76,30-160 +device,string,no,Device that logged the data,Fitbit, +quality_flag,string,no,Data quality status,ok,ok|suspect|missing diff --git a/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_wearable-checkpoint.csv b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_wearable-checkpoint.csv new file mode 100644 index 00000000..17067984 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/.ipynb_checkpoints/schema_wearable-checkpoint.csv @@ -0,0 +1,11 @@ +field_name,type,required,description,example,range +patient_id,string,yes,Unique ID for the patient,P001, +timestamp,datetime,yes,Time recorded in UTC,2025-08-22T09:30:00Z, +step_count,int,no,Steps counted in interval,120,0-50000 +activity_level,string,no,Activity intensity,light,sedentary|light|moderate|intense +accel_x,float,no,Acceleration X-axis,0.12, +accel_y,float,no,Acceleration Y-axis,-0.03, +accel_z,float,no,Acceleration Z-axis,9.81, +calories,float,no,Calories burned in interval,4.5,0-10000 +device,string,no,Device that logged the data,Apple Watch, +quality_flag,string,no,Data quality status,ok,ok|suspect|missing diff --git a/AI Guardian/DataSchemas_WJ/raw_data/sample_emotion.csv b/AI Guardian/DataSchemas_WJ/raw_data/sample_emotion.csv new file mode 100644 index 00000000..5d2b2ef6 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/sample_emotion.csv @@ -0,0 +1,11 @@ +patient_id,timestamp,source,emotion,confidence,notes,model,privacy_tag +P001,2000-01-01T09:00:00Z,face,neutral,0.91,,v1,normal +P001,2000-01-01T09:05:00Z,voice,stressed,0.77,,v1,normal +P001,2000-01-01T09:10:00Z,face,happy,0.85,,v1,normal +P001,2000-01-01T09:15:00Z,voice,tired,0.66,,v1,normal +P001,2000-01-01T09:20:00Z,face,angry,0.72,,v1,normal +P002,2000-01-01T09:25:00Z,voice,neutral,0.88,,v1,normal +P002,2000-01-01T09:30:00Z,face,surprise,0.79,,v1,normal +P002,2000-01-01T09:35:00Z,voice,sad,0.81,,v1,normal +P002,2000-01-01T09:40:00Z,face,happy,0.92,,v1,normal +P002,2000-01-01T09:45:00Z,voice,stressed,0.84,,v1,normal diff --git a/AI Guardian/DataSchemas_WJ/raw_data/sample_vital.csv b/AI Guardian/DataSchemas_WJ/raw_data/sample_vital.csv new file mode 100644 index 00000000..f61f0507 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/sample_vital.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,heart_rate,spo2,resp_rate,temp,systolic_bp,diastolic_bp,device,quality_flag +P001,2000-01-01T09:00:00Z,86,96,17,36.8,125,82,Fitbit,ok +P001,2000-01-01T09:05:00Z,91,98,15,36.6,117,85,Fitbit,ok +P001,2000-01-01T09:10:00Z,77,95,18,37.0,138,73,Fitbit,ok +P001,2000-01-01T09:15:00Z,101,99,16,36.9,133,78,Fitbit,ok +P001,2000-01-01T09:20:00Z,84,97,14,36.5,121,87,Fitbit,ok +P001,2000-01-01T09:25:00Z,92,96,19,37.2,130,74,Fitbit,ok +P001,2000-01-01T09:30:00Z,79,98,13,36.7,116,81,Fitbit,ok +P001,2000-01-01T09:35:00Z,88,94,15,36.9,127,75,Fitbit,ok +P001,2000-01-01T09:40:00Z,95,97,17,36.6,119,84,Fitbit,ok +P001,2000-01-01T09:45:00Z,83,99,16,36.8,122,80,Fitbit,ok +P002,2000-01-01T09:00:00Z,74,97,16,36.7,121,78,Apple Watch,ok +P002,2000-01-01T09:05:00Z,88,96,14,36.8,118,82,Apple Watch,ok +P002,2000-01-01T09:10:00Z,81,95,17,36.6,125,85,Apple Watch,ok +P002,2000-01-01T09:15:00Z,90,98,15,36.9,119,76,Apple Watch,ok +P002,2000-01-01T09:20:00Z,85,99,18,36.7,123,80,Apple Watch,ok +P002,2000-01-01T09:25:00Z,93,96,16,36.8,127,83,Apple Watch,ok +P002,2000-01-01T09:30:00Z,76,94,14,36.5,122,79,Apple Watch,ok +P002,2000-01-01T09:35:00Z,89,97,17,36.9,124,81,Apple Watch,ok +P002,2000-01-01T09:40:00Z,82,98,15,36.7,120,77,Apple Watch,ok +P002,2000-01-01T09:45:00Z,87,95,16,36.6,126,84,Apple Watch,ok diff --git a/AI Guardian/DataSchemas_WJ/raw_data/sample_wearable.csv b/AI Guardian/DataSchemas_WJ/raw_data/sample_wearable.csv new file mode 100644 index 00000000..1b3822bb --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/sample_wearable.csv @@ -0,0 +1,21 @@ +patient_id,timestamp,step_count,activity_level,accel_x,accel_y,accel_z,calories,device,quality_flag +P001,2000-01-01T09:00:00Z,0,sedentary,0.05,-0.01,9.77,1.5,Fitbit,ok +P001,2000-01-01T09:05:00Z,45,light,0.10,0.03,9.85,2.3,Fitbit,ok +P001,2000-01-01T09:10:00Z,120,moderate,0.08,0.02,9.70,3.8,Fitbit,ok +P001,2000-01-01T09:15:00Z,230,intense,-0.12,0.04,9.92,5.6,Fitbit,ok +P001,2000-01-01T09:20:00Z,60,light,0.02,-0.05,9.81,2.7,Fitbit,ok +P001,2000-01-01T09:25:00Z,180,moderate,0.06,0.01,9.76,4.9,Fitbit,ok +P001,2000-01-01T09:30:00Z,15,sedentary,-0.04,0.02,9.79,1.8,Fitbit,ok +P001,2000-01-01T09:35:00Z,95,light,0.09,-0.03,9.88,3.2,Fitbit,ok +P001,2000-01-01T09:40:00Z,310,intense,0.14,0.05,9.83,6.4,Fitbit,ok +P001,2000-01-01T09:45:00Z,150,moderate,-0.08,0.06,9.74,4.2,Fitbit,ok +P002,2000-01-01T09:00:00Z,20,light,0.04,0.01,9.79,2.0,Apple Watch,ok +P002,2000-01-01T09:05:00Z,80,moderate,-0.06,0.02,9.82,3.1,Apple Watch,ok +P002,2000-01-01T09:10:00Z,200,intense,0.11,-0.04,9.75,6.0,Apple Watch,ok +P002,2000-01-01T09:15:00Z,35,light,0.02,0.00,9.80,2.4,Apple Watch,ok +P002,2000-01-01T09:20:00Z,145,moderate,-0.05,0.03,9.84,4.7,Apple Watch,ok +P002,2000-01-01T09:25:00Z,250,intense,0.13,-0.02,9.77,7.2,Apple Watch,ok +P002,2000-01-01T09:30:00Z,10,sedentary,0.00,0.01,9.81,1.6,Apple Watch,ok +P002,2000-01-01T09:35:00Z,95,light,0.07,-0.01,9.85,3.3,Apple Watch,ok +P002,2000-01-01T09:40:00Z,310,intense,-0.09,0.04,9.78,6.5,Apple Watch,ok +P002,2000-01-01T09:45:00Z,175,moderate,0.05,0.06,9.82,4.8,Apple Watch,ok diff --git a/AI Guardian/DataSchemas_WJ/raw_data/schema_alert.csv b/AI Guardian/DataSchemas_WJ/raw_data/schema_alert.csv new file mode 100644 index 00000000..7a13ddb6 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/schema_alert.csv @@ -0,0 +1,10 @@ +field_name,type,required,description,example,range +alert_id,string,yes,Unique alert ID,A-0001, +patient_id,string,yes,Patient this alert is about,P001, +timestamp,datetime,yes,Alert creation time,2025-08-22T10:00:00Z, +alert_type,string,yes,Type of alert,tachycardia,tachycardia|hypoxia|fever|bradycardia|hypertension|hypotension|emotion_risk +severity,string,yes,How serious it is,medium,low|medium|high|critical +trigger_fields,string,no,Fields that triggered alert,"heart_rate,spo2", +rule_id,string,no,Rule that caused alert,R-HEART-01, +ack_status,string,no,If alert has been acknowledged,unread,unread|acknowledged|resolved +notes,string,no,Free text for reviewer,Short run post-exercise, diff --git a/AI Guardian/DataSchemas_WJ/raw_data/schema_vital.csv b/AI Guardian/DataSchemas_WJ/raw_data/schema_vital.csv new file mode 100644 index 00000000..30e1ed53 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/schema_vital.csv @@ -0,0 +1,11 @@ +field_name,type,required,description,example,range +patient_id,string,yes,Unique ID for the patient,P001, +timestamp,datetime,yes,Time recorded in UTC,2025-08-22T09:30:00Z, +heart_rate,int,yes,Beats per minute,72,30-220 +spo2,int,yes,Oxygen % level,98,70-100 +resp_rate,int,no,Breaths per min,16,4-60 +temp,float,no,Body temp in Celsius,36.8,30-43 +systolic_bp,int,no,Top blood pressure value,118,60-260 +diastolic_bp,int,no,Bottom blood pressure value,76,30-160 +device,string,no,Device that logged the data,Fitbit, +quality_flag,string,no,Data quality status,ok,ok|suspect|missing diff --git a/AI Guardian/DataSchemas_WJ/raw_data/schema_wearable.csv b/AI Guardian/DataSchemas_WJ/raw_data/schema_wearable.csv new file mode 100644 index 00000000..17067984 --- /dev/null +++ b/AI Guardian/DataSchemas_WJ/raw_data/schema_wearable.csv @@ -0,0 +1,11 @@ +field_name,type,required,description,example,range +patient_id,string,yes,Unique ID for the patient,P001, +timestamp,datetime,yes,Time recorded in UTC,2025-08-22T09:30:00Z, +step_count,int,no,Steps counted in interval,120,0-50000 +activity_level,string,no,Activity intensity,light,sedentary|light|moderate|intense +accel_x,float,no,Acceleration X-axis,0.12, +accel_y,float,no,Acceleration Y-axis,-0.03, +accel_z,float,no,Acceleration Z-axis,9.81, +calories,float,no,Calories burned in interval,4.5,0-10000 +device,string,no,Device that logged the data,Apple Watch, +quality_flag,string,no,Data quality status,ok,ok|suspect|missing