Upload 5 files

.gitignore

@@ -0,0 +1,11 @@
+# Model weights
+*.pth
+
+# Optional model folder
+models/
+
+# Python cache
+__pycache__/
+*.pyc
+
+# Jupyter checkpoints

app.py

@@ -0,0 +1,612 @@
+import streamlit as st
+import torch
+import torch.nn as nn
+from torchvision import models, transforms
+from PIL import Image
+import numpy as np
+
+# ── Page configuration ──────────────────────────────────
+st.set_page_config(
+    page_title="DermaScan AI",
+    page_icon="🔬",
+    layout="wide",
+    initial_sidebar_state="expanded"
+)
+
+# ── Custom CSS styling ──────────────────────────────────
+st.markdown("""
+<style>
+    @import url('https://fonts.googleapis.com/css2?family=Syne:wght@400;600;700;800&family=DM+Sans:ital,opsz,wght@0,9..40,300;0,9..40,400;0,9..40,500;1,9..40,300&display=swap');
+
+    :root {
+        --bg-base:       #0A0E1A;
+        --bg-surface:    #111827;
+        --bg-elevated:   #1A2235;
+        --border:        rgba(99,179,237,0.12);
+        --border-bright: rgba(99,179,237,0.35);
+        --accent-cyan:   #63B3ED;
+        --accent-teal:   #4FD1C5;
+        --accent-green:  #68D391;
+        --accent-amber:  #F6AD55;
+        --accent-red:    #FC8181;
+        --text-primary:  #EDF2F7;
+        --text-secondary:#A0AEC0;
+        --text-muted:    #4A5568;
+    }
+
+    html, body, [data-testid="stAppViewContainer"] {
+        background: var(--bg-base) !important;
+        font-family: 'DM Sans', sans-serif;
+        color: var(--text-primary);
+    }
+
+    [data-testid="stSidebar"] {
+        background: var(--bg-surface) !important;
+        border-right: 1px solid var(--border) !important;
+    }
+
+    [data-testid="stSidebar"] * {
+        color: var(--text-primary) !important;
+    }
+
+    /* Hide default streamlit chrome */
+    #MainMenu, footer, header { visibility: hidden; }
+    [data-testid="stDecoration"] { display: none; }
+
+    /* ── Hero banner ── */
+    .hero {
+        display: flex;
+        align-items: center;
+        gap: 1.5rem;
+        padding: 2.2rem 2.5rem;
+        background: linear-gradient(135deg, #0D1B2E 0%, #112240 60%, #0D2137 100%);
+        border: 1px solid var(--border-bright);
+        border-radius: 16px;
+        margin-bottom: 2rem;
+        position: relative;
+        overflow: hidden;
+    }
+    .hero::before {
+        content: '';
+        position: absolute;
+        top: -40px; right: -40px;
+        width: 220px; height: 220px;
+        background: radial-gradient(circle, rgba(99,179,237,0.12) 0%, transparent 70%);
+        border-radius: 50%;
+    }
+    .hero-icon {
+        font-size: 3rem;
+        line-height: 1;
+        filter: drop-shadow(0 0 12px rgba(99,179,237,0.5));
+    }
+    .hero-text h1 {
+        font-family: 'Syne', sans-serif;
+        font-size: 2rem;
+        font-weight: 800;
+        color: var(--text-primary);
+        margin: 0 0 0.2rem 0;
+        letter-spacing: -0.5px;
+    }
+    .hero-text p {
+        font-size: 0.95rem;
+        color: var(--text-secondary);
+        margin: 0;
+        font-weight: 300;
+        letter-spacing: 0.02em;
+    }
+    .hero-badge {
+        margin-left: auto;
+        background: rgba(99,179,237,0.1);
+        border: 1px solid var(--border-bright);
+        color: var(--accent-cyan);
+        padding: 0.4rem 1rem;
+        border-radius: 20px;
+        font-size: 0.78rem;
+        font-weight: 600;
+        letter-spacing: 0.08em;
+        text-transform: uppercase;
+        white-space: nowrap;
+    }
+
+    /* ── Upload zone ── */
+    .upload-label {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.75rem;
+        font-weight: 700;
+        letter-spacing: 0.12em;
+        text-transform: uppercase;
+        color: var(--accent-cyan);
+        margin-bottom: 0.5rem;
+    }
+    [data-testid="stFileUploader"] {
+        background: var(--bg-elevated) !important;
+        border: 1.5px dashed var(--border-bright) !important;
+        border-radius: 12px !important;
+        padding: 0.5rem !important;
+        transition: border-color 0.2s;
+    }
+    [data-testid="stFileUploader"]:hover {
+        border-color: var(--accent-cyan) !important;
+    }
+    [data-testid="stFileUploader"] * { color: var(--text-secondary) !important; }
+
+    /* ── Section headers ── */
+    .section-header {
+        display: flex;
+        align-items: center;
+        gap: 0.6rem;
+        margin-bottom: 1rem;
+    }
+    .section-header span.icon { font-size: 1rem; }
+    .section-header span.label {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.72rem;
+        font-weight: 700;
+        letter-spacing: 0.13em;
+        text-transform: uppercase;
+        color: var(--accent-cyan);
+    }
+    .section-divider {
+        flex: 1;
+        height: 1px;
+        background: var(--border);
+        margin-left: 0.5rem;
+    }
+
+    /* ── Image panel ── */
+    .img-panel {
+        background: var(--bg-surface);
+        border: 1px solid var(--border);
+        border-radius: 14px;
+        padding: 1.2rem;
+        overflow: hidden;
+    }
+    .img-panel img {
+        border-radius: 10px !important;
+        width: 100% !important;
+    }
+    .img-meta {
+        display: flex;
+        justify-content: space-between;
+        margin-top: 0.8rem;
+        padding-top: 0.8rem;
+        border-top: 1px solid var(--border);
+    }
+    .img-meta-item {
+        text-align: center;
+    }
+    .img-meta-item .val {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.9rem;
+        font-weight: 700;
+        color: var(--text-primary);
+    }
+    .img-meta-item .key {
+        font-size: 0.7rem;
+        color: var(--text-muted);
+        text-transform: uppercase;
+        letter-spacing: 0.08em;
+    }
+
+    /* ── Primary diagnosis card ── */
+    .diagnosis-card {
+        background: linear-gradient(135deg, #0D1B2E 0%, #0D2137 100%);
+        border: 1px solid var(--border-bright);
+        border-radius: 14px;
+        padding: 1.5rem 1.8rem;
+        margin-bottom: 1.2rem;
+        position: relative;
+        overflow: hidden;
+    }
+    .diagnosis-card::after {
+        content: '';
+        position: absolute;
+        bottom: -30px; right: -30px;
+        width: 120px; height: 120px;
+        background: radial-gradient(circle, rgba(79,209,197,0.1) 0%, transparent 70%);
+        border-radius: 50%;
+    }
+    .diagnosis-label {
+        font-size: 0.68rem;
+        font-weight: 700;
+        letter-spacing: 0.14em;
+        text-transform: uppercase;
+        color: var(--text-muted);
+        margin-bottom: 0.4rem;
+    }
+    .diagnosis-name {
+        font-family: 'Syne', sans-serif;
+        font-size: 1.5rem;
+        font-weight: 800;
+        color: var(--text-primary);
+        margin-bottom: 0.8rem;
+        line-height: 1.2;
+    }
+    .confidence-pill {
+        display: inline-flex;
+        align-items: center;
+        gap: 0.4rem;
+        padding: 0.35rem 0.85rem;
+        border-radius: 20px;
+        font-size: 0.82rem;
+        font-weight: 600;
+    }
+    .pill-high   { background: rgba(104,211,145,0.15); color: var(--accent-green); border: 1px solid rgba(104,211,145,0.3); }
+    .pill-medium { background: rgba(246,173,85,0.15);  color: var(--accent-amber); border: 1px solid rgba(246,173,85,0.3); }
+    .pill-low    { background: rgba(252,129,129,0.15); color: var(--accent-red);   border: 1px solid rgba(252,129,129,0.3); }
+    .pill-dot {
+        width: 6px; height: 6px;
+        border-radius: 50%;
+        background: currentColor;
+    }
+
+    /* ── Prediction rows ── */
+    .pred-row {
+        background: var(--bg-elevated);
+        border: 1px solid var(--border);
+        border-radius: 10px;
+        padding: 0.9rem 1.2rem;
+        margin-bottom: 0.6rem;
+        transition: border-color 0.2s;
+    }
+    .pred-row:hover { border-color: var(--border-bright); }
+    .pred-row-top {
+        display: flex;
+        justify-content: space-between;
+        align-items: center;
+        margin-bottom: 0.5rem;
+    }
+    .pred-rank {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.65rem;
+        font-weight: 700;
+        color: var(--text-muted);
+        letter-spacing: 0.1em;
+    }
+    .pred-name {
+        font-weight: 500;
+        color: var(--text-primary);
+        font-size: 0.92rem;
+    }
+    .pred-pct {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.88rem;
+        font-weight: 700;
+        color: var(--accent-cyan);
+    }
+    .pred-bar-track {
+        height: 4px;
+        background: var(--bg-base);
+        border-radius: 4px;
+        overflow: hidden;
+    }
+    .pred-bar-fill {
+        height: 100%;
+        border-radius: 4px;
+        background: linear-gradient(90deg, var(--accent-cyan) 0%, var(--accent-teal) 100%);
+        transition: width 0.6s ease;
+    }
+
+    /* ── Clinical note ── */
+    .clinical-note {
+        background: rgba(99,179,237,0.06);
+        border: 1px solid rgba(99,179,237,0.2);
+        border-left: 3px solid var(--accent-cyan);
+        border-radius: 10px;
+        padding: 1rem 1.2rem;
+        margin-top: 1.5rem;
+    }
+    .clinical-note .cn-title {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.7rem;
+        font-weight: 700;
+        letter-spacing: 0.12em;
+        text-transform: uppercase;
+        color: var(--accent-cyan);
+        margin-bottom: 0.3rem;
+    }
+    .clinical-note p {
+        font-size: 0.83rem;
+        color: var(--text-secondary);
+        margin: 0;
+        line-height: 1.6;
+    }
+
+    /* ── Sidebar styles ── */
+    .sidebar-section-title {
+        font-family: 'Syne', sans-serif;
+        font-size: 0.68rem;
+        font-weight: 700;
+        letter-spacing: 0.13em;
+        text-transform: uppercase;
+        color: #63B3ED !important;
+        margin-bottom: 0.6rem;
+    }
+    .sidebar-info {
+        background: rgba(99,179,237,0.07) !important;
+        border: 1px solid rgba(99,179,237,0.18) !important;
+        border-radius: 10px !important;
+        padding: 0.9rem !important;
+        font-size: 0.83rem !important;
+        color: #A0AEC0 !important;
+        line-height: 1.6 !important;
+    }
+    .model-spec-row {
+        display: flex;
+        justify-content: space-between;
+        padding: 0.45rem 0;
+        border-bottom: 1px solid rgba(99,179,237,0.1);
+        font-size: 0.82rem;
+    }
+    .model-spec-row:last-child { border-bottom: none; }
+    .spec-key { color: #4A5568; }
+    .spec-val { color: #EDF2F7; font-weight: 500; }
+
+    /* ── Empty state ── */
+    .empty-state {
+        text-align: center;
+        padding: 4rem 2rem;
+        background: var(--bg-surface);
+        border: 1.5px dashed var(--border);
+        border-radius: 16px;
+        margin-top: 1rem;
+    }
+    .empty-state .es-icon { font-size: 3.5rem; margin-bottom: 1rem; opacity: 0.5; }
+    .empty-state h3 {
+        font-family: 'Syne', sans-serif;
+        color: var(--text-secondary);
+        font-size: 1.2rem;
+        margin-bottom: 0.5rem;
+    }
+    .empty-state p { color: var(--text-muted); font-size: 0.9rem; }
+
+    /* ── Streamlit overrides ── */
+    [data-testid="stMarkdownContainer"] p { color: var(--text-secondary); }
+    .stSpinner > div { color: var(--accent-cyan) !important; }
+    [data-testid="stDataFrame"] {
+        background: var(--bg-surface) !important;
+        border: 1px solid var(--border) !important;
+        border-radius: 10px !important;
+    }
+    [data-testid="stProgress"] > div > div > div > div {
+        background: linear-gradient(90deg, #63B3ED, #4FD1C5) !important;
+        border-radius: 4px;
+    }
+    [data-testid="stInfo"] {
+        background: rgba(99,179,237,0.07) !important;
+        border: 1px solid rgba(99,179,237,0.2) !important;
+        color: var(--text-secondary) !important;
+        border-radius: 10px !important;
+    }
+</style>
+""", unsafe_allow_html=True)
+
+# ── Load model ──────────────────────────────────────────
+@st.cache_resource
+def load_model():
+    classes = np.load('classes.npy', allow_pickle=True)
+    num_classes = len(classes)
+    model = models.resnet50(weights=None)
+    num_features = model.fc.in_features
+    model.fc = nn.Sequential(
+        nn.Linear(num_features, 1024), nn.ReLU(), nn.Dropout(0.2),
+        nn.Linear(1024, 512),          nn.ReLU(), nn.Dropout(0.1),
+        nn.Linear(512, 128),           nn.ReLU(),
+        nn.Linear(128, num_classes)
+    )
+    model.load_state_dict(torch.load('skin_lesion_resnet50_best.pth', map_location='cpu'))
+    model.eval()
+    return model, classes
+
+def preprocess(image):
+    transform = transforms.Compose([
+        transforms.Resize((224, 224)),
+        transforms.ToTensor(),
+        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+    ])
+    return transform(image).unsqueeze(0)
+
+def confidence_pill(conf):
+    if conf >= 0.80:
+        return f'<span class="confidence-pill pill-high"><span class="pill-dot"></span>{conf*100:.1f}% High Confidence</span>'
+    elif conf >= 0.60:
+        return f'<span class="confidence-pill pill-medium"><span class="pill-dot"></span>{conf*100:.1f}% Moderate</span>'
+    else:
+        return f'<span class="confidence-pill pill-low"><span class="pill-dot"></span>{conf*100:.1f}% Low Confidence</span>'
+
+# ── Hero ─────────────────────────────────────────────────
+st.markdown("""
+<div class="hero">
+    <div class="hero-icon">🔬</div>
+    <div class="hero-text">
+        <h1>DermaScan AI</h1>
+        <p>Deep learning–powered dermoscopy analysis &nbsp;·&nbsp; ResNet50 classification engine</p>
+    </div>
+    <div class="hero-badge">⚡ Live Inference</div>
+</div>
+""", unsafe_allow_html=True)
+
+# ── Sidebar ──────────────────────────────────────────────
+with st.sidebar:
+    st.markdown('<p class="sidebar-section-title">About</p>', unsafe_allow_html=True)
+    st.markdown("""
+    <div class="sidebar-info">
+    DermaScan AI uses a fine-tuned <strong style="color:#EDF2F7">ResNet50</strong> convolutional network
+    trained on a comprehensive dermoscopy dataset to classify skin lesion types from uploaded images.
+    </div>
+    """, unsafe_allow_html=True)
+
+    st.markdown("<br>", unsafe_allow_html=True)
+    st.markdown('<p class="sidebar-section-title">Model Specs</p>', unsafe_allow_html=True)
+    st.markdown("""
+    <div style="background:rgba(99,179,237,0.05); border:1px solid rgba(99,179,237,0.15); border-radius:10px; padding:0.7rem 1rem;">
+        <div class="model-spec-row"><span class="spec-key">Architecture</span><span class="spec-val">ResNet50</span></div>
+        <div class="model-spec-row"><span class="spec-key">Framework</span><span class="spec-val">PyTorch</span></div>
+        <div class="model-spec-row"><span class="spec-key">Input Size</span><span class="spec-val">224 × 224 px</span></div>
+        <div class="model-spec-row"><span class="spec-key">Task</span><span class="spec-val">Multi-class</span></div>
+        <div class="model-spec-row"><span class="spec-key">Normalization</span><span class="spec-val">ImageNet</span></div>
+    </div>
+    """, unsafe_allow_html=True)
+
+    st.markdown("<br>", unsafe_allow_html=True)
+    st.markdown('<p class="sidebar-section-title">Instructions</p>', unsafe_allow_html=True)
+    st.markdown("""
+    <div style="font-size:0.82rem; color:#718096; line-height:1.8;">
+    1. Upload a dermoscopy image (JPG / PNG)<br>
+    2. Wait for model inference to complete<br>
+    3. Review ranked predictions and confidence scores<br>
+    4. Consult a dermatologist for clinical decisions
+    </div>
+    """, unsafe_allow_html=True)
+
+# ── Upload ────────────────────────────────────────────────
+st.markdown("""
+<div class="section-header">
+    <span class="icon">📂</span>
+    <span class="label">Image Upload</span>
+    <div class="section-divider"></div>
+</div>
+""", unsafe_allow_html=True)
+
+uploaded_file = st.file_uploader(
+    "Drop a dermoscopy image here, or click to browse",
+    type=["jpg", "jpeg", "png"],
+    help="Supported: JPG, JPEG, PNG"
+)
+
+# ── Results ────────────────────────────────────────────────
+if uploaded_file:
+    image = Image.open(uploaded_file).convert("RGB")
+    w, h = image.size
+
+    col1, col2 = st.columns([1, 1], gap="large")
+
+    with col1:
+        st.markdown("""
+        <div class="section-header" style="margin-top:1.2rem">
+            <span class="icon">🖼</span>
+            <span class="label">Input Image</span>
+            <div class="section-divider"></div>
+        </div>
+        """, unsafe_allow_html=True)
+
+        st.markdown('<div class="img-panel">', unsafe_allow_html=True)
+        st.image(image, use_column_width=True)
+        st.markdown(f"""
+        <div class="img-meta">
+            <div class="img-meta-item">
+                <div class="val">{w} × {h}</div>
+                <div class="key">Resolution</div>
+            </div>
+            <div class="img-meta-item">
+                <div class="val">{uploaded_file.name.split('.')[-1].upper()}</div>
+                <div class="key">Format</div>
+            </div>
+            <div class="img-meta-item">
+                <div class="val">{uploaded_file.size // 1024} KB</div>
+                <div class="key">File Size</div>
+            </div>
+            <div class="img-meta-item">
+                <div class="val">RGB</div>
+                <div class="key">Color Mode</div>
+            </div>
+        </div>
+        </div>
+        """, unsafe_allow_html=True)
+
+    with col2:
+        st.markdown("""
+        <div class="section-header" style="margin-top:1.2rem">
+            <span class="icon">🎯</span>
+            <span class="label">Analysis Results</span>
+            <div class="section-divider"></div>
+        </div>
+        """, unsafe_allow_html=True)
+
+        with st.spinner("Running model inference…"):
+            model, classes = load_model()
+            tensor = preprocess(image)
+            with torch.no_grad():
+                outputs = model(tensor)
+                probs = torch.softmax(outputs, dim=1)[0]
+                top_prob, top_idx = torch.topk(probs, 3)
+
+        top_prediction = classes[top_idx[0]]
+        top_confidence = float(top_prob[0])
+
+        # Primary diagnosis
+        st.markdown(f"""
+        <div class="diagnosis-card">
+            <div class="diagnosis-label">Primary Diagnosis</div>
+            <div class="diagnosis-name">{top_prediction}</div>
+            {confidence_pill(top_confidence)}
+        </div>
+        """, unsafe_allow_html=True)
+
+        # Top-3 predictions
+        st.markdown("""
+        <div class="section-header" style="margin-top:1rem">
+            <span class="icon">📊</span>
+            <span class="label">Ranked Predictions</span>
+            <div class="section-divider"></div>
+        </div>
+        """, unsafe_allow_html=True)
+
+        rank_labels = ["1st", "2nd", "3rd"]
+        for i, (prob, idx) in enumerate(zip(top_prob, top_idx)):
+            conf = float(prob)
+            bar_width = int(conf * 100)
+            st.markdown(f"""
+            <div class="pred-row">
+                <div class="pred-row-top">
+                    <div>
+                        <span class="pred-rank">{rank_labels[i]}&nbsp;&nbsp;</span>
+                        <span class="pred-name">{classes[idx]}</span>
+                    </div>
+                    <span class="pred-pct">{conf*100:.1f}%</span>
+                </div>
+                <div class="pred-bar-track">
+                    <div class="pred-bar-fill" style="width:{bar_width}%"></div>
+                </div>
+            </div>
+            """, unsafe_allow_html=True)
+
+    # ── Full results table ──
+    st.markdown("<br>", unsafe_allow_html=True)
+    st.markdown("""
+    <div class="section-header">
+        <span class="icon">📋</span>
+        <span class="label">Classification Summary</span>
+        <div class="section-divider"></div>
+    </div>
+    """, unsafe_allow_html=True)
+
+    results_data = {
+        "Rank":       [f"#{i+1}" for i in range(len(top_prob))],
+        "Diagnosis":  [classes[idx] for idx in top_idx],
+        "Confidence": [f"{prob*100:.2f}%" for prob in top_prob],
+        "Status":     [
+            "✅ Primary" if i == 0 else ("⚠️ Alternate" if i == 1 else "ℹ️ Low prob")
+            for i in range(len(top_prob))
+        ]
+    }
+    st.dataframe(results_data, use_container_width=True, hide_index=True)
+
+    # ── Clinical note ──
+    st.markdown("""
+    <div class="clinical-note">
+        <div class="cn-title">⚕️ Clinical Disclaimer</div>
+        <p>This AI classification is intended for <strong style="color:#EDF2F7">informational and research purposes only</strong>.
+        It does not constitute a medical diagnosis. Always consult a board-certified dermatologist
+        for clinical evaluation, diagnosis, and treatment recommendations.</p>
+    </div>
+    """, unsafe_allow_html=True)
+
+else:
+    st.markdown("""
+    <div class="empty-state">
+        <div class="es-icon">🔬</div>
+        <h3>No Image Uploaded</h3>
+        <p>Upload a dermoscopy image above to begin skin lesion classification</p>
+    </div>
+    """, unsafe_allow_html=True)

classes.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:591d581f6d5af2122399c75052b256dbf06ee4945254890509b8182866a29a52
+size 320

skin_lesion_resnet50_best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:036d9e997e9fc17e9a746fb0125914c91f55bd2ae477f2739e5cbeab07de8685
+size 105116363