Update app.py

app.py

@@ -1,24 +1,61 @@
 # app.py
 import math
-import numpy as np
+import tempfile
+from pathlib import Path
 import gradio as gr
 
 # ==========================================
 # CARPL Multi-Use-Case ROI Calculators (MMG / FFR-CT / MSK AI)
 # Shared UI/UX: Inter font, white cards, pill header, Overall + Tabs + Waterfall + Evidence + CTA
+# Extras: card-style use case chooser, MMG vendor presets, CSV export (fixed)
 # ==========================================
 
 USE_CASES = ["Mammography AI (MMG)", "FFR-CT AI", "MSK AI (ER/Trauma)"]
-CTA_URL = "https://carpl.ai/contact-us"  # swap to HubSpot/Calendly if you want
+CTA_URL = "https://carpl.ai/contact-us"  # swap to your HubSpot/Calendly
 CTA_LABEL = "Book a 15-min walkthrough"
 
+MMG_VENDOR_PRESETS = {
+    "Custom": {},
+    "Lunit": {
+        "base_recall_rate": 0.028, "ai_recall_rate": 0.025,
+        "base_ppr": 0.100, "ai_ppr": 0.095,
+        "ai_audit_rate": 0.050, "base_audit_rate": 0.000,
+        "read_reduction_pct": 0.15,
+        "followup_uplift_pct": 0.0095,
+        "early_detect_uplift_per_1000": 0.7,
+    },
+    "Therapixel (MammoScreen)": {
+        "base_recall_rate": 0.028, "ai_recall_rate": 0.024,
+        "base_ppr": 0.100, "ai_ppr": 0.094,
+        "ai_audit_rate": 0.040, "base_audit_rate": 0.000,
+        "read_reduction_pct": 0.15,
+        "followup_uplift_pct": 0.010,
+        "early_detect_uplift_per_1000": 0.9,
+    },
+    "MammoScreen (Alt)": {
+        "base_recall_rate": 0.030, "ai_recall_rate": 0.026,
+        "base_ppr": 0.100, "ai_ppr": 0.093,
+        "ai_audit_rate": 0.045, "base_audit_rate": 0.000,
+        "read_reduction_pct": 0.18,
+        "followup_uplift_pct": 0.011,
+        "early_detect_uplift_per_1000": 0.8,
+    },
+    "MedCognetics": {
+        "base_recall_rate": 0.029, "ai_recall_rate": 0.026,
+        "base_ppr": 0.100, "ai_ppr": 0.096,
+        "ai_audit_rate": 0.035, "base_audit_rate": 0.000,
+        "read_reduction_pct": 0.14,
+        "followup_uplift_pct": 0.009,
+        "early_detect_uplift_per_1000": 0.6,
+    },
+}
+
 # ---------- Helpers ----------
 def usd(x: float, digits: int = 0) -> str:
     if x == math.inf:
         return "∞"
     try:
-        fmt = f"{{x:,.{digits}f}}"
-        return "$" + fmt.format(x=x)
+        return "$" + f"{x:,.{digits}f}"
     except Exception:
         return "$0"
 
@@ -31,8 +68,17 @@ def pct(x: float, digits: int = 1) -> str:
 def clamp_nonneg(x: float) -> float:
     return max(0.0, float(x))
 
-def label_pct(delta: float) -> str:
-    return f"{max(0.0, delta)*100:.1f}%"
+def write_csv(rows, title: str = "roi_results") -> str:
+    """Write rows [(label, value), ...] to a temp CSV and return file path."""
+    csv_dir = Path(tempfile.gettempdir())
+    path = csv_dir / f"{title.replace(' ','_').lower()}.csv"
+    with open(path, "w", encoding="utf-8") as f:
+        f.write("Metric,Value\n")
+        for lab, val in rows:
+            # strip any HTML tags if present
+            val = str(val).replace("<b>", "").replace("</b>", "")
+            f.write(f"\"{lab}\",\"{val}\"\n")
+    return str(path)
 
 # ---------- MMG (Mammography) ----------
 def compute_mmg(
@@ -55,17 +101,14 @@ def compute_mmg(
     integration_overhead_monthly: float,
     cloud_compute_monthly: float,
 ):
-    # Uptake fixed to 100% for MMG (can be toggled in backend if needed)
     monthly_ai_cases = clamp_nonneg(monthly_volume)
     annual_ai_cases = monthly_ai_cases * 12.0
 
-    # Clinical deltas
     errors_reduced     = clamp_nonneg(monthly_ai_cases * (base_ppr - ai_ppr))
     discrepant_flags   = clamp_nonneg(monthly_ai_cases * (ai_audit_rate - base_audit_rate))
     recalls_avoided    = clamp_nonneg(monthly_ai_cases * (base_recall_rate - ai_recall_rate))
     earlier_detections = clamp_nonneg(monthly_ai_cases * (early_detect_uplift_per_1000 / 1000.0))
 
-    # Operational
     base_read_seconds = read_minutes * 60.0
     hours_saved = clamp_nonneg(monthly_ai_cases * (base_read_seconds * read_reduction_pct) / 3600.0)
     workload_reduction_pct = read_reduction_pct
@@ -73,7 +116,6 @@ def compute_mmg(
     capacity_increase_pct = (1.0 / max(1e-6, (1.0 - read_reduction_pct)) - 1.0)
     value_time_saved_month = hours_saved * radiologist_hourly_cost
 
-    # Financial
     baseline_monthly_cost         = monthly_ai_cases * base_cost_per_scan
     new_monthly_cost              = baseline_monthly_cost * (1.0 - cost_reduction_pct)
     per_scan_cost_savings_month   = baseline_monthly_cost - new_monthly_cost
@@ -82,7 +124,6 @@ def compute_mmg(
     recall_cost_savings_month     = recalls_avoided * recall_cost_per_case
     early_detection_savings_month = earlier_detections * treatment_cost_delta_early_vs_late
 
-    # Program costs (backend)
     vendor_cost_month   = monthly_ai_cases * vendor_per_case_fee
     platform_cost_month = platform_annual_fee / 12.0
     other_costs_month   = integration_overhead_monthly + cloud_compute_monthly
@@ -92,16 +133,12 @@ def compute_mmg(
     ops_value_month    = value_time_saved_month + per_scan_cost_savings_month + recall_cost_savings_month + early_detection_savings_month
 
     net_impact_month = incr_revenue_month - incr_costs_month + ops_value_month
-    roi_pct_val      = (net_impact_month / incr_costs_month) if incr_costs_month > 0 else float("nan")
-
-    # Annual KPIs
-    net_impact_annual = net_impact_month * 12.0
-    roi_pct_annual    = (net_impact_annual / (incr_costs_month*12.0)) if incr_costs_month > 0 else float("nan")
-    annual_program_cost = platform_annual_fee + vendor_per_case_fee*annual_ai_cases + other_costs_month*12.0
-    net_impact_per_ai_case_month = net_impact_month / max(1.0, monthly_ai_cases)
-    months_to_payback = (annual_program_cost / max(1e-6, net_impact_per_ai_case_month)) / max(1.0, monthly_ai_cases)
+    roi_pct_annual   = ((net_impact_month*12) / (incr_costs_month*12)) if incr_costs_month > 0 else float("nan")
+    months_to_payback = (
+        (platform_annual_fee + vendor_per_case_fee*annual_ai_cases + other_costs_month*12.0)
+        / max(1e-6, (net_impact_month / max(1.0, monthly_ai_cases))) / max(1.0, monthly_ai_cases)
+    )
 
-    # Evidence snapshot (neutral placeholders)
     evidence = """
     <ul class='evidence'>
       <li>Modeled reductions in recalls and false positives reduce unnecessary follow-ups and costs.</li>
@@ -110,20 +147,18 @@ def compute_mmg(
     </ul>
     """
 
-    # Clinical bullet for one-liner
     clinical_bullet = (
         f"~{int(round(recalls_avoided))} recalls avoided, "
         f"{earlier_detections:.1f} earlier cancers detected, "
         f"{int(round(errors_reduced))} fewer missed positives"
     )
 
-    # Pack the standard structure
     return {
         "summary": f"For your practice with {int(monthly_volume):,} mammography scans/month, modeled net benefit is {usd(net_impact_month)} per month. Clinical: {clinical_bullet}.",
         "financial": {
             "rows": [
                 ("Additional follow-up scans (count/mo)", f"{int(round(addl_followups))}"),
-                ("Additional follow-up revenue (mo)", usd(incr_revenue_month)),
+                ("Additional follow-up revenue (mo)", usd(addl_followup_revenue_month)),
                 ("Value of time saved (mo)", usd(value_time_saved_month)),
                 ("Per-scan radiologist cost savings (mo)", usd(per_scan_cost_savings_month)),
                 ("Savings from avoided recalls (mo)", usd(recall_cost_savings_month)),
@@ -132,7 +167,7 @@ def compute_mmg(
                 ("Platform license (mo)", usd(platform_cost_month)),
                 ("Integration & cloud (mo)", usd(other_costs_month)),
                 ("Net impact (mo)", f"<b>{usd(net_impact_month)}</b>"),
-                ("Net impact (annual)", f"<b>{usd(net_impact_annual)}</b>"),
+                ("Net impact (annual)", f"<b>{usd(net_impact_month*12)}</b>"),
                 ("ROI % (annual)", f"<b>{'' if math.isnan(roi_pct_annual) else f'{roi_pct_annual*100:.1f}%'}</b>"),
                 ("Months to payback", f"<b>{months_to_payback:.1f}</b>"),
             ]
@@ -157,12 +192,12 @@ def compute_mmg(
                 ("Effective capacity increase", pct(capacity_increase_pct)),
             ]
         },
-        "waterfall": [("Incremental revenue", incr_revenue_month), ("Incremental costs", -incr_costs_month), ("Operational value", ops_value_month)],
+        "waterfall": [("Incremental revenue", addl_followup_revenue_month), ("Incremental costs", - (vendor_cost_month + platform_cost_month + other_costs_month)), ("Operational value", ops_value_month)],
         "annual_card": {
-            "incr_rev": incr_revenue_month * 12.0,
-            "incr_costs": incr_costs_month * 12.0,
+            "incr_rev": addl_followup_revenue_month * 12.0,
+            "incr_costs": (vendor_cost_month + platform_cost_month + other_costs_month) * 12.0,
             "ops_value": ops_value_month * 12.0,
-            "net": net_impact_annual,
+            "net": net_impact_month * 12.0,
             "roi_pct": roi_pct_annual,
             "payback": months_to_payback,
         },
@@ -176,16 +211,13 @@ def compute_ffrct(
     uptake_pct: float,
     avg_time_to_decision_today_hours: float,
     baseline_clinician_touch_min: float,
-    # Assumptions (kept visible for clarity; feel free to tuck into gr.State if you prefer)
     reimb_ccta: float, reimb_ffrct: float, reimb_ai_qpa: float, pct_billed_ai_qpa: float,
     one_test_dx_pct: float, dec_unnec_ica_pct: float, more_likely_revasc_pct: float, revasc_prevalence_pct: float,
     vendor_per_case_cost: float, platform_annual_cost: float, stress_test_cost: float,
     bed_hour_value: float, clinician_hour_cost: float, ai_time_to_decision_min: float,
     clinician_touch_reduction_pct: float, baseline_diag_ica_rate_pct: float, baseline_additional_testing_rate_pct: float,
-    # Sensitivity
     sens_uptake_factor_pct: float, sens_dec_unnec_ica_factor_pct: float, sens_vendor_cost_factor_pct: float,
 ):
-    # Site-specific net cost per diagnostic ICA (rough anchors; adjust as needed)
     if site_type == "Hospital / Health System":
         net_cost_per_diag_ica = 5000.0
     elif site_type == "Imaging Center":
@@ -198,18 +230,16 @@ def compute_ffrct(
     annual_eligible = monthly_eligible_ccta * 12.0
     annual_uptake_cases = annual_eligible * uptake
 
-    # Normalize %
-    pct_ai_qpa = max(0.0, min(1.0, pct_billed_ai_qpa/100.0))
-    one_test_dx = max(0.0, min(1.0, one_test_dx_pct/100.0))
+    pct_ai_qpa     = max(0.0, min(1.0, pct_billed_ai_qpa/100.0))
+    one_test_dx    = max(0.0, min(1.0, one_test_dx_pct/100.0))
     need_addl_with_ai = 1.0 - one_test_dx
-    dec_unnec_ica = max(0.0, min(1.0, dec_unnec_ica_pct/100.0 * max(0.0, min(1.0, sens_dec_unnec_ica_factor_pct/100.0))))
+    dec_unnec_ica  = max(0.0, min(1.0, dec_unnec_ica_pct/100.0 * max(0.0, min(1.0, sens_dec_unnec_ica_factor_pct/100.0))))
     more_likely_revasc = max(0.0, min(1.0, more_likely_revasc_pct/100.0))
-    revasc_prev = max(0.0, min(1.0, revasc_prevalence_pct/100.0))
-    vendor_cost = float(vendor_per_case_cost) * max(0.0, min(1.0, sens_vendor_cost_factor_pct/100.0))
+    revasc_prev    = max(0.0, min(1.0, revasc_prevalence_pct/100.0))
+    vendor_cost    = float(vendor_per_case_cost) * max(0.0, min(1.0, sens_vendor_cost_factor_pct/100.0))
     platform_annual_cost = float(platform_annual_cost)
-    stress_test_cost = float(stress_test_cost)
+    stress_test_cost     = float(stress_test_cost)
 
-    # Baseline vs With-AI
     baseline_revenue = annual_eligible * reimb_ccta
     baseline_additional_tests = annual_eligible * max(0.0, min(1.0, baseline_additional_testing_rate_pct/100.0))
     baseline_additional_tests_cost = baseline_additional_tests * stress_test_cost
@@ -220,7 +250,6 @@ def compute_ffrct(
     baseline_ops_value = 0.0
     baseline_costs = baseline_additional_tests_cost + baseline_unnecessary_ica_cost
 
-    # With AI
     with_ai_revenue = (
         annual_eligible * reimb_ccta
         + annual_uptake_cases * reimb_ffrct
@@ -230,7 +259,7 @@ def compute_ffrct(
     with_ai_platform_costs = platform_annual_cost
 
     baseline_addl_tests_in_ai_cohort = annual_uptake_cases * max(0.0, min(1.0, baseline_additional_testing_rate_pct/100.0))
-    with_ai_additional_tests = annual_uptake_cases * need_addl_with_ai
+    with_ai_additional_tests = annual_uptake_cases * (1.0 - one_test_dx)
     with_ai_additional_tests_cost = with_ai_additional_tests * stress_test_cost
     avoided_additional_tests = max(0.0, baseline_addl_tests_in_ai_cohort - with_ai_additional_tests)
 
@@ -238,7 +267,6 @@ def compute_ffrct(
     with_ai_unnecessary_ica = max(0.0, baseline_unnecessary_ica - avoided_unnec_ica)
     with_ai_unnecessary_ica_cost = with_ai_unnecessary_ica * net_cost_per_diag_ica
 
-    # Ops value
     ai_saved_hours_per_case = min(max(0.0, ai_time_to_decision_min/60.0), max(0.0, float(avg_time_to_decision_today_hours)))
     bed_hours_saved = annual_uptake_cases * ai_saved_hours_per_case
     bed_hours_value = bed_hours_saved * bed_hour_value
@@ -247,9 +275,6 @@ def compute_ffrct(
     with_ai_ops_value = bed_hours_value + clinician_hours_value
 
     with_ai_costs = with_ai_vendor_costs + with_ai_platform_costs + with_ai_additional_tests_cost + with_ai_unnecessary_ica_cost
-    with_ai_net = with_ai_revenue - with_ai_costs + with_ai_ops_value
-
-    # Incremental
     incr_revenue = with_ai_revenue - baseline_revenue
     incr_costs   = with_ai_costs - baseline_costs
     incr_ops     = with_ai_ops_value - baseline_ops_value
@@ -260,7 +285,7 @@ def compute_ffrct(
 
     per_case_net_impact = (net_impact / annual_uptake_cases) if annual_uptake_cases > 0 else 0.0
     cases_to_payback = (ai_program_costs / per_case_net_impact) if per_case_net_impact > 0 else math.inf
-    months_to_payback = ( (cases_to_payback / (monthly_eligible_ccta * (uptake))) if (monthly_eligible_ccta * uptake) > 0 and cases_to_payback != math.inf else math.inf )
+    months_to_payback = ((cases_to_payback / (monthly_eligible_ccta * (uptake))) if (monthly_eligible_ccta * uptake) > 0 and cases_to_payback != math.inf else math.inf)
 
     evidence = """
     <ul class='evidence'>
@@ -316,26 +341,21 @@ def compute_ffrct(
         "evidence": evidence,
     }
 
-# ---------- MSK (ER/Trauma) ----------
+# ---------- MSK ----------
 def compute_msk(
     scans_per_day: float,
     reading_time_min: float,
     er_time_to_treatment_min: float,
     radiologist_hourly_cost: float = 180.0,
 ):
-    # Port of your MSK calc logic into our standard structure
     scans_per_month = clamp_nonneg(scans_per_day) * 30.0
-    # Clinical
-    errors_reduced_per_month   = int(scans_per_month * 0.05 * 0.20)  # example: 5% incidence, 20% improvement
+    errors_reduced_per_month   = int(scans_per_month * 0.05 * 0.20)
     discrepant_cases_flagged   = int(scans_per_month * 0.05)
-    # Operational
     hrs_saved_per_visit        = max(0.0, er_time_to_treatment_min) * 0.50 / 60.0
     time_saved_per_scan_min    = max(0.0, reading_time_min) * 0.30
     total_time_saved_hours     = (scans_per_month * time_saved_per_scan_min) / 60.0
     value_time_saved_month     = total_time_saved_hours * radiologist_hourly_cost
-    # Financial (simple proxy)
-    radiologist_cost_savings   = scans_per_month * 4.0  # your original per-scan proxy
-    # Net (no vendor/platform modeled here; keep backend if you want later)
+    radiologist_cost_savings   = scans_per_month * 4.0
     incr_revenue_month = 0.0
     incr_costs_month   = 0.0
     ops_value_month    = value_time_saved_month + radiologist_cost_savings
@@ -364,24 +384,11 @@ def compute_msk(
                 ("Discrepant cases flagged (est.)", f"{discrepant_cases_flagged} /mo"),
                 ("Hours saved per ED visit (modeled)", f"{hrs_saved_per_visit:.2f}"),
             ],
-            "bars": [
-                ("Touch-time reduction", 0.30),
-            ]
-        },
-        "operational": {
-            "rows": [
-                ("Radiologist hours saved / month", f"{total_time_saved_hours:.1f}"),
-            ]
-        },
-        "waterfall": [("Incremental revenue", incr_revenue_month), ("Incremental costs", -incr_costs_month), ("Operational value", ops_value_month)],
-        "annual_card": {
-            "incr_rev": incr_revenue_month * 12.0,
-            "incr_costs": incr_costs_month * 12.0,
-            "ops_value": ops_value_month * 12.0,
-            "net": net_impact_month * 12.0,
-            "roi_pct": float("nan"),
-            "payback": float("nan"),
+            "bars": [("Touch-time reduction", 0.30)]
         },
+        "operational": {"rows": [("Radiologist hours saved / month", f"{total_time_saved_hours:.1f}")]},
+        "waterfall": [("Incremental revenue", 0.0), ("Incremental costs", 0.0), ("Operational value", ops_value_month)],
+        "annual_card": {"incr_rev": 0.0, "incr_costs": 0.0, "ops_value": ops_value_month * 12.0, "net": net_impact_month * 12.0, "roi_pct": float("nan"), "payback": float("nan")},
         "evidence": evidence,
     }
 
@@ -407,11 +414,11 @@ def build_overall_card(title: str, summary_line: str, annual):
     </div>
     """
 
-def build_rows_card(title: str, rows: list) -> str:
+def build_rows_card(title: str, rows):
     items = "".join(f"<div>{lab}</div><div>{val}</div>" for lab, val in rows)
     return f"<div class='card'><div style='font-weight:700;margin-bottom:6px'>{title}</div><div class='kpi-grid'>{items}</div></div>"
 
-def build_clinical_card(rows: list, bars: list) -> str:
+def build_clinical_card(rows, bars):
     rows_html = "".join(f"<div>{lab}</div><div>{val}</div>" for lab, val in rows)
     bars_html = ""
     for lab, frac in (bars or []):
@@ -424,7 +431,7 @@ def build_clinical_card(rows: list, bars: list) -> str:
           </div>"""
     return f"<div class='card'><div style='font-weight:700;margin-bottom:6px'>Clinical</div><div class='kpi-grid'>{rows_html}</div><div class='bars'>{bars_html}</div></div>"
 
-def build_waterfall(wf_rows: list, title="Waterfall (monthly)") -> str:
+def build_waterfall(wf_rows, title="Waterfall (monthly)"):
     denom = sum(abs(v) for _, v in wf_rows) or 1.0
     def row(label, val):
         width = min(100, max(2, int(abs(val)/denom * 100)))
@@ -434,6 +441,10 @@ def build_waterfall(wf_rows: list, title="Waterfall (monthly)") -> str:
         title, "".join(row(l, v) for l, v in wf_rows), usd(sum(v for _, v in wf_rows))
     )
 
+def rows_to_csv(fin_rows, clin_rows, op_rows) -> str:
+    all_rows = [("Section","—")] + [("Financial","—")] + fin_rows + [("Clinical","—")] + clin_rows + [("Operational","—")] + op_rows
+    return write_csv(all_rows, title="roi_results")
+
 # ---------- UI ----------
 def build_ui():
     with gr.Blocks(theme=gr.themes.Soft(), css="""
@@ -461,6 +472,32 @@ def build_ui():
 .header{display:flex;justify-content:space-between;align-items:center}
 .header .title{font-weight:800;font-size:1.1rem}
 .header .pill{margin-left:8px}
+
+/* Use-case chooser as cards */
+#uc [role="radiogroup"] {
+  display: grid;
+  grid-template-columns: repeat(3, minmax(0,1fr));
+  gap: 12px;
+}
+@media (max-width: 820px) {
+  #uc [role="radiogroup"] { grid-template-columns: 1fr; }
+}
+#uc [role="radiogroup"] label {
+  border: 1px solid #eef2f7;
+  border-radius: 16px;
+  background: #fff;
+  padding: 14px 16px;
+  box-shadow: 0 8px 24px rgba(0,0,0,.06);
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+  gap: 10px;
+  transition: transform .06s ease, box-shadow .12s ease, border-color .12s ease;
+}
+#uc [role="radiogroup"] input[type="radio"] { accent-color: #14b8a6; }
+#uc [role="radiogroup"] label:hover { transform: translateY(-1px); box-shadow: 0 12px 28px rgba(0,0,0,.10); }
+#uc [role="radiogroup"] input[type="radio"]:checked + span { font-weight: 700; }
+#uc .uc-hint { margin-top: 6px; opacity: .7; font-size: .9em; }
 """) as demo:
         gr.Markdown("""
 <div class='header'>
@@ -471,9 +508,29 @@ def build_ui():
 
         with gr.Row():
             with gr.Column(scale=1):
-                use_case = gr.Dropdown(USE_CASES, value=USE_CASES[0], label="Use case")
+                # Use case chooser (as cards)
+                with gr.Column(elem_id="uc"):
+                    gr.Markdown("### Choose your use case")
+                    use_case = gr.Radio(
+                        choices=[
+                            "🩺 Mammography AI (MMG)",
+                            "❤️‍🩹 FFR-CT AI",
+                            "🦴 MSK AI (ER/Trauma)"
+                        ],
+                        value="🩺 Mammography AI (MMG)",
+                        label=None,
+                        interactive=True
+                    )
+                    gr.Markdown("<div class='uc-hint'>Pick one to load the matching calculator. You can switch anytime.</div>")
 
                 # --- MMG Inputs ---
+                vendor_preset = gr.Dropdown(
+                    choices=list(MMG_VENDOR_PRESETS.keys()),
+                    value="Custom",
+                    label="MMG vendor preset",
+                    info="Prefills hidden sensitivity. You can still override in Sensitivity.",
+                    visible=True
+                )
                 mmg_monthly_volume = gr.Slider(0, 20000, 7500, step=50, label="Monthly volume (MMG)", info="Mammography exams per month", visible=True)
                 mmg_read_minutes   = gr.Number(label="Avg reading time today (minutes)", value=1.7, visible=True)
                 mmg_rdx_hr_cost    = gr.Number(label="Radiologist cost (USD/hr)", value=180, visible=True)
@@ -493,7 +550,6 @@ def build_ui():
                     mmg_follow_uplift = gr.Slider(0, 0.2, value=0.0095, step=0.0005, label="Follow-up uplift (fraction of AI scans)")
                     mmg_early_uplift_per_1000 = gr.Number(value=0.7, label="Earlier cancers detected (+/1000 AI scans)")
                     mmg_tx_delta = gr.Number(value=15000.0, label="Treatment cost savings per earlier case (USD)")
-                # Hidden program costs
                 mmg_vendor_fee = gr.State(2.5)
                 mmg_platform_annual = gr.State(12000.0)
                 mmg_integration_mo  = gr.State(0.0)
@@ -547,19 +603,26 @@ def build_ui():
                 waterfall_panel = gr.HTML()
                 evidence_panel  = gr.HTML()
                 cta_panel       = gr.HTML(visible=False)
+                csv_file        = gr.File(label="Download CSV", visible=False)
 
-        # Toggle visibility by use case
-        def _on_use_case_change(uc):
+        def normalize_uc(uclabel: str) -> str:
+            if "Mammography" in uclabel: return USE_CASES[0]
+            if "FFR-CT" in uclabel: return USE_CASES[1]
+            return USE_CASES[2]
+
+        def _on_use_case_change(uc_label):
+            uc = normalize_uc(uc_label)
             mmg_vis = (uc == USE_CASES[0])
             f_vis   = (uc == USE_CASES[1])
             msk_vis = (uc == USE_CASES[2])
             return (
-                gr.update(visible=mmg_vis), gr.update(visible=mmg_vis), gr.update(visible=mmg_vis), gr.update(visible=mmg_vis),
-                gr.update(visible=mmg_vis),
+                gr.update(visible=mmg_vis),  # vendor preset
+                gr.update(visible=mmg_vis), gr.update(visible=mmg_vis), gr.update(visible=mmg_vis),
+                gr.update(visible=mmg_vis),  # accordion
 
                 gr.update(visible=f_vis), gr.update(visible=f_vis), gr.update(visible=f_vis), gr.update(visible=f_vis), gr.update(visible=f_vis),
                 gr.update(visible=f_vis),  # accordion
-                # MSK
+
                 gr.update(visible=msk_vis), gr.update(visible=msk_vis), gr.update(visible=msk_vis),
             )
 
@@ -567,15 +630,37 @@ def build_ui():
             _on_use_case_change,
             inputs=[use_case],
             outputs=[
-                mmg_monthly_volume, mmg_read_minutes, mmg_rdx_hr_cost, mmg_sens, mmg_sens,  # (accordion doubled on purpose to keep visibility synced)
+                vendor_preset,
+                mmg_monthly_volume, mmg_read_minutes, mmg_rdx_hr_cost, mmg_sens,
                 f_site, f_monthly_eligible, f_uptake, f_ttd_hours, f_touch_min, f_assump,
                 msk_scans_day, msk_read_min, msk_er_ttt_min
             ],
         )
 
-        # Dispatcher
+        def _apply_vendor_preset(preset_name,
+                                 base_ppr, ai_ppr, base_audit, ai_audit, base_recall, ai_recall,
+                                 read_redux, follow_uplift, early_uplift):
+            p = MMG_VENDOR_PRESETS.get(preset_name, {})
+            return (
+                gr.update(value=p.get("base_ppr", base_ppr)),
+                gr.update(value=p.get("ai_ppr", ai_ppr)),
+                gr.update(value=p.get("base_audit_rate", base_audit)),
+                gr.update(value=p.get("ai_audit_rate", ai_audit)),
+                gr.update(value=p.get("base_recall_rate", base_recall)),
+                gr.update(value=p.get("ai_recall_rate", ai_recall)),
+                gr.update(value=p.get("read_reduction_pct", read_redux)),
+                gr.update(value=p.get("followup_uplift_pct", follow_uplift)),
+                gr.update(value=p.get("early_detect_uplift_per_1000", early_uplift)),
+            )
+
+        vendor_preset.change(
+            _apply_vendor_preset,
+            inputs=[vendor_preset, mmg_base_ppr, mmg_ai_ppr, mmg_base_audit, mmg_ai_audit, mmg_base_recall, mmg_ai_recall, mmg_read_redux, mmg_follow_uplift, mmg_early_uplift_per_1000],
+            outputs=[mmg_base_ppr, mmg_ai_ppr, mmg_base_audit, mmg_ai_audit, mmg_base_recall, mmg_ai_recall, mmg_read_redux, mmg_follow_uplift, mmg_early_uplift_per_1000],
+        )
+
         def _compute(
-            uc,
+            uclabel,
             # MMG
             mv, rm, rhr,
             base_ppr, ai_ppr, base_audit, ai_audit, base_recall, ai_recall, recall_cost, read_redux, cps, cps_redux, fol_price, fol_uplift, early_uplift, tx_delta,
@@ -586,6 +671,8 @@ def build_ui():
             # MSK
             msk_day, msk_read, msk_ttt, msk_rhr,
         ):
+            uc = normalize_uc(uclabel)
+
             if uc == USE_CASES[0]:
                 res = compute_mmg(
                     mv, rm, rhr,
@@ -612,9 +699,15 @@ def build_ui():
             operational_html = build_rows_card("Operational", res["operational"]["rows"])
             water = build_waterfall(res["waterfall"])
             evidence = f"<div class='card'><div style='font-weight:700;margin-bottom:6px'>Evidence snapshot</div>{res['evidence']}<div class='small-note'>Neutral claims; update with site citations.</div></div>"
-            cta_line = "Book a quick walkthrough with our team."
-            cta = f"<div class='card cta'><div>{cta_line}</div><a class='cta-btn' href='{CTA_URL}' target='_blank' rel='noopener'>{CTA_LABEL}</a></div>"
-            return overall_html, financial_html, clinical_html, operational_html, water, evidence, gr.update(value=cta, visible=True)
+            cta = f"<div class='card cta'><div>Want to see this in your workflow?</div><a class='cta-btn' href='{CTA_URL}' target='_blank' rel='noopener'>{CTA_LABEL}</a></div>"
+
+            # CSV export (flatten current rows) -> write to temp file
+            fin_rows = [(lab, val) for lab, val in res["financial"]["rows"]]
+            clin_rows = [(lab, val) for lab, val in res["clinical"]["rows"]]
+            op_rows   = [(lab, val) for lab, val in res["operational"]["rows"]]
+            csv_path = rows_to_csv(fin_rows, clin_rows, op_rows)
+
+            return overall_html, financial_html, clinical_html, operational_html, water, evidence, gr.update(value=cta, visible=True), gr.update(value=csv_path, visible=True)
 
         inputs = [
             use_case,
@@ -628,7 +721,7 @@ def build_ui():
             # MSK inputs
             msk_scans_day, msk_read_min, msk_er_ttt_min, msk_rdx_hr_cost,
         ]
-        outputs = [overall_card, financial_card, clinical_card, operational_card, waterfall_panel, evidence_panel, cta_panel]
+        outputs = [overall_card, financial_card, clinical_card, operational_card, waterfall_panel, evidence_panel, cta_panel, csv_file]
 
         run_btn.click(_compute, inputs=inputs, outputs=outputs)
         demo.load(_compute, inputs=inputs, outputs=outputs)