Homework_2

[The-Paul2002]

🏥 Homework Assignment: Geospatial Analysis of Hospitals in Peru

---

0) Environment & Reproducibility

For environment setup and reproducibility, please use the same model and setup provided in this repository:
👉 Rodrigo Grijalba – Python Dashboard Class

Follow the same structure for creating and managing your environment (requirements.txt, Streamlit, GeoPandas, Folium, etc.).

---

1) Repository Setup

• Create a new public GitHub repository named Hospitals-Access-Peru.
• Use the Rodrigo John Paradox repository as a template.

---

2) Geospatial Analysis with GeoPandas

📥 Datasets & Official Links

• Hospitals (MINSA – IPRESS)
  👉 MINSA – Registro Nacional de IPRESS

  • Use only the operational subset: filter records to hospitals that are operational (“functioning status”).
  • Clarification: “Functioning status” refers to the official MINSA database of operational/active hospitals, according to the dataset and data dictionary.
  • Keep only records with valid coordinates (latitude and longitude).

• Population Centers (INEI)
  👉 Dataset – Centros Poblados

• Administrative Boundaries (Districts of Peru)
  Use the district shapefile employed in this reference class:
  👉 Class: GeoPandas – Districts of Peru
  Recommended CRS: EPSG:4326.

Key columns (hospitals):
INSTITUCIÓN, NOMBRE DEL ESTABLECIMIENTO, CLASIFICACION, DEPARTAMENTO, PROVINCIA, DISTRITO, ESTADO ,LATITUD (NORTE) ,LONGITUD (ESTE)

---

🗺️ Task 1: Static Maps — Hospital Count by District

• Load the district shapefile of Peru (link above).
• With GeoPandas, create static maps showing the number of operational public hospitals per district:
  • Map 1: Total public hospitals per district.
  • Map 2: Highlight districts with zero hospitals.
  • Map 3: Top 10 districts with the highest number of hospitals (distinct color scale).

---

📊 Task 2: Department-level Analysis

• Aggregate at the department level.
• Calculate the total number of operational hospitals in each department.
• Identify:
  • The department with the highest number of hospitals.
  • The department with the lowest number of hospitals.
• Present the results with:
  • A summary table (sorted from highest to lowest).
  • A bar chart (matplotlib or seaborn).
  • A department-level choropleth map (GeoPandas).

---

📍 Task 3: Proximity Analysis (using Population Centers)

Regions to analyze:

• Lima
• Loreto

Methodology:

1. From Population Centers, compute the centroid of each locality (or use the provided geometry centroids).
2. For each centroid, calculate the number of operational hospitals within a 10 km buffer.
3. Identify (per region):
   • The population center with the fewest hospitals nearby (isolation).
   • The population center with the most hospitals nearby (concentration).
4. Plot with folium:
   • The selected population center (centroid).
   • The 10 km buffer (circle).
   • All hospitals inside the buffer.

Notes:

• Ensure all layers use CRS EPSG:4326 (or project to EPSG:32718 for metric 10 km buffers and then reproject back to EPSG:4326 for visualization).
• Document any cleaning steps (duplicate removal, invalid coordinates).

---

3) Interactive Mapping with Folium

🗺️ Task 1: National Choropleth (District Level)

• Build a Folium choropleth of the number of hospitals per district.
• Add a marker cluster with all hospital points.

---

📍 Task 2: Proximity Visualization — Lima & Loreto

• Reproduce the proximity analysis in Folium:
  • Red circle (10 km): population center with the lowest hospital density nearby.
  • Green circle (10 km): population center with the highest hospital density nearby.
• Add popups/tooltips with population center name and number of hospitals inside the radius.

Short written analysis:

• Lima: urban concentration and accessibility.
• Loreto: geographic dispersion and accessibility challenges in the Amazon.

---

4) Application Deployment with Streamlit

The Streamlit application must be divided into 3 parts (tabs), following the template shown in class (see example image with icons and labels):

• 🗂️ Tab 1: Data Description

  • Unit of analysis: operational public hospitals in Peru.
  • Data sources: MINSA – IPRESS (operational subset), Population Centers.
  • Filtering rules: only operational hospitals with valid lat/long.

• 🗺️ Tab 2: Static Maps & Department Analysis

  • Embed static maps created with GeoPandas.
  • Include the department summary table and bar chart.

• 🌍 Tab 3: Dynamic Maps

  • National Folium choropleth + markers.
  • Folium proximity maps for Lima & Loreto.

(otherwise they will get points deducted)
Example:

Examples Ouputs

Data Desctription

Maps Statics

Maps Dinamycs

---

📌 Notes

• Use clean and reproducible code (functions, notebooks, or scripts).
• Comment CRS conversions and data processing steps.
• Include a Makefile or simple run instructions (optional).
• Document in the README how hospitals were filtered for “functioning status”.

---

📅 Deadline

September 18 2025

---

📊 Example Outputs

• Data Description: data analysis of hospitals in Peru.
• Static Maps: choropleth (counts), zero-hospital districts, top-10 districts.
• Dynamic Maps: national choropleth + markers; proximity maps (Lima = high density, Loreto = low density).

---

📤 Submission (Repository & Dashboard Links)

Please submit both your:

1. GitHub repository URL, and
2. Deployed Streamlit dashboard URL

in the following Google Sheet:
👉 Submission Excel – Repository & Dashboard Links

Your Streamlit app must be deployed (e.g., Streamlit Community Cloud) and linked in the sheet above.