dkguo · esump25 · Jul 15, 2024 · Jul 15, 2024 · Jul 15, 2024 · Jul 15, 2024
diff --git a/Copy_of_Carmen_CNNs.ipynb b/Copy_of_Carmen_CNNs.ipynb
diff --git a/IMG_2469.jpg b/IMG_2469.jpg
diff --git a/Main_CNNs.ipynb b/Main_CNNs.ipynb
diff --git a/contact_us.png b/contact_us.png
diff --git a/dataset_example.png b/dataset_example.png
diff --git a/eye-retina-disease-classification.code-workspace b/eye-retina-disease-classification.code-workspace
@@ -0,0 +1,11 @@
+{
+	"folders": [
+		{
+			"path": "."
+		},
+		{
+			"path": "../../../Downloads/AIS Group IV"
+		}
+	],
+	"settings": {}
+}
diff --git a/index.html b/index.html
@@ -15,71 +15,98 @@
 ==============================================================================
 -->
 
-<!doctype html>
-
+<!DOCTYPE html>
+<html lang="en">
 <head>
-  <title>Project Classification+X</title>
-  <meta charset="UTF-8">
-  <meta name="viewport" content="width=device-width, initial-scale=1">
-  <link rel="stylesheet" href="shared/tfjs-examples.css" />
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>AI Scholars: Retinal Disease Classification</title>
+    <style>
+      body {
+          font-family: Arial, sans-serif;
+          line-height: 1.6;
+          margin: 0; /* Remove default margin */
+          padding: 0;
+          display: flex;
+          flex-direction: column;
+          min-height: 100vh; /* Ensure the body takes up at least the full viewport height */
+      }
+      header {
+          background: #500101;
+          color: #fff;
+          padding: 10px 0;
+          text-align: center;
+
+      }
+      .logo {
+          width: 250px; /* Adjust logo size as needed */
+          height: auto;
+          margin-right: 10px;
+      }
+      .container {
+          flex: 1; /* Make the container grow to fill remaining space */
+          max-width: 800px;
+          margin: auto;
+          padding: 20px;
+      }
+      .main-content {
+            margin-bottom: 20px;
+            margin-right: 10px;
+            text-align: left;
+      }
+      footer {
+          background: #500101;
+          color: #fff;
+          text-align: center;
+          padding: 10px 0;
+          width: 100%;
+      }
+  </style>
 </head>
-
-<style>
-  .pred-container {
-    margin-bottom: 20px;
-  }
-
-  .pred-container > div {
-    display: inline-block;
-    margin-right: 20px;
-    vertical-align: top;
-  }
-
-  .row {
-    display: table-row;
-  }
-  .cell {
-    display: table-cell;
-    padding-right: 20px;
-  }
-
-  #file-container {
-    margin-bottom: 20px;
-  }
-</style>
-
 <body>
-  <div class="tfjs-example-container">
-    <section class='title-area'>
-      <h1>Fruits and Veggies Classification</h1>
-    </section>
-
-    <section>
-      <img src="dataset.jpg" style="width: 50%;">
-      <p class='section-head'>Description</p>
-      <p>
-        We classify 9 categories of fruits and veggies. 
-      </p>
-    </section>
-
-    <section>
-      <p class='section-head'>Status</p>
-      <div id="status"></div>
-    </section>
-
-    <section>
-      <p class='section-head'>Model Output</p>
-
-      <div id="file-container" style="display: none">
-        Upload an image: <input type="file" id="files" name="files[]" multiple />
-      </div>
-
-      <div id="predictions"></div>
-
-      <img style="display: none" id="cat" src="ai4all.jpg" width=224 height=224 />
-    </section>
-
-    <script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@1.5.2"></script>
-    <script src="index.js"></script>
-  </div>
+    <header>
+        <img src="https://media.licdn.com/dms/image/C5622AQEifm61PFunnA/feedshare-shrink_2048_1536/0/1660748532397?e=2147483647&v=beta&t=0aEC7AeQ6gZ7AIKyllhnx4A_Ry3bQ4GWyx5PTeaJwjk" alt="Logo" class="logo" width="200" height="60" >
+        <h1>AI Scholars: Eye Retina Disease Classification</h1>
+    </header>
+    <div class="container">
+        <section class="main-content">
+            <h2>About</h2>
+            <p>Throughout the AI Scholars program, a group of 5 students created an AI image classification model to determine eye diseases using a dataset of retina scans. Alongside professors and a graduate student at Carnegie Mellon University, we created a goal in striving to provide cheap and accessible medical exams in areas where healthcare is limited. We coded an AI model in an attempt to solve problems of inaccessibility and the risk of death through eye diseases. 1.1 billion people in the world suffer from visual impairments, and retinal diseases are the leading cause for blindness in low-income countries. There’s only 200,000 Ophthalmologists worldwide compared to its population of over 8 billion people.</p>
+        </section>
+        <section>
+            <h2>Model Card</h2>
+            <img src="model_card.png" width="900" height="450">
+        </section>
+        <section>
+          <h2>Sample Image</h2>
+          <img src="dataset_example.png">
+        </section>
+        <section>
+          <h2>AI Model</h2>
+          <h2><a href="https://colab.research.google.com/drive/1yfQusHB9gilfSgGEteRLz7BIUBdIeYe-?usp=sharing">https://colab.research.google.com/drive/1yfQusHB9gilfSgGEteRLz7BIUBdIeYe-?usp=sharing</a></h2>
+        </section>
+        <section>
+          <h2>How the Model Works</h2>
+          <p> The dataset we trained our models with is called the Retinal Fundus Multi-Disease Image Dataset. We trained our models with 1,920 images, split 80:20 between the training images and validation images, and tested with 640 unseen images. The dataset was collected at Nanded, India with retinal fundus cameras. We use python convolutional neural networks (CNN), to run through the training data repeatedly and come up with the optimal values of each weight and bias. They use the highlighted areas of the filtered images to determine the weight those specific pixels need to have. We ran our data through both binary and multi-classification models, determining if they are diseased or not diseased, while also moving on to determining which disease it is. By training our dataset, we received varying levels of accuracy based on the hyperparameters we used. Although it is still an ongoing training process, our model is able to learn how to classify different retina scans and its diseases.
+           </p>
+        </section>
+        <section>
+          <h2><a href="https://aisgroup4retinaldisease.my.canva.site">Presentation Slideshow</a></h2>
+        </section>
+        <section>
+          <h2>Contact Us</h2>
+          <img src="contact_us.png" width="900" height="450">
+          <p>Malik Traore -- malikt@andrew.cmu.edu</p>
+          <p>Carmen Ung -- cung@andrew.cmu.edu</p>
+          <p>Emmy Sumpter -- esumpter@andrew.cmu.edu</p>
+          <p>Tommy Weng -- tjweng@andrew.cmu.edu</p>
+          <p>Juan Cruz Field -- jcfield@andrew.cmu.edu</p>
+          <p>Peya Mowar -- pmowar@andrew.cmu.edu</p>
+        </section>
+
+    </div>
+    <footer>
+        <p>&copy; 2024 CMU Artificial Intelligence Scholars</p>
+    </footer>
 </body>
+</html>
diff --git a/main.py b/main.py
diff --git a/model_card.png b/model_card.png
diff --git a/model_tfjs/group1-shard1of4.bin b/model_tfjs/group1-shard1of4.bin
diff --git a/model_tfjs/group1-shard2of4.bin b/model_tfjs/group1-shard2of4.bin
diff --git a/model_tfjs/group1-shard3of4.bin b/model_tfjs/group1-shard3of4.bin
diff --git a/model_tfjs/group1-shard4of4.bin b/model_tfjs/group1-shard4of4.bin
diff --git a/model_tfjs/model.json b/model_tfjs/model.json
diff --git a/style.css b/style.css
@@ -0,0 +1 @@
+