Upload app.py with huggingface_hub

app.py

@@ -1,58 +1,101 @@
 import gradio as gr
-import os
+import os, requests, glob
 from groq import Groq
 
-GROQ_KEY = os.environ.get("GROQ_API_KEY", "")
+GROQ_KEY = os.environ.get("GROQ_API_KEY","")
 client = Groq(api_key=GROQ_KEY)
 
 KNOWHOW = """
-CardioLab SJSU Know-How:
-MCL: Sylgard 184 at 10:1 ratio, 48hr cure, green laser PIV, 70bpm 5L/min flow
-TGT: Arduino Uno + Stepper Motor, 150mL blood, sample at 0/20/40/60min, measure TAT PF1.2 hemolysis platelets
-uPAD: Jaffe reaction creatinine + picric acid = orange-red, normal <1.2 mg/dL, CKD >1.5 mg/dL
-FSI: COMSOL ALE mesh, blood density 1060 kg/m3, viscosity 0.0035 Pa.s, St Jude geometry
-MHV: 27mm SJM Regent, bileaflet, trileaflet, monoleaflet, pediatric designs
-CKD: Stages 1-5 by creatinine, uPAD image RGB analysis 64x64 pixel
+SJSU CardioLab Know-How:
+MCL: Mock Circulatory Loop using Sylgard 184 PDMS at 10:1 ratio cured 48hrs, green laser PIV, 70bpm 5L/min physiological flow
+TGT: Thrombogenicity Tester with Arduino Uno + Stepper Motor, 150mL blood, sample at 0/20/40/60min, measure TAT PF1.2 hemolysis platelets
+uPAD: Microfluidic Paper Device, Jaffe reaction creatinine + picric acid = orange-red color, normal creatinine 0.6-1.2 mg/dL, CKD above 1.5 mg/dL
+FSI: COMSOL Multiphysics ALE mesh, blood density 1060 kg/m3, viscosity 0.0035 Pa.s, St Jude Medical geometry reference
+MHV: 27mm SJM Regent bileaflet mechanical heart valve, also trileaflet, monoleaflet, pediatric designs studied
+CKD Stages: Stage 1 below 1.5, Stage 2 1.5-3.0, Stage 3-4 3.0-6.0, Stage 5 above 6.0 mg/dL creatinine
+Equipment: Heska Element HT5 hematology analyzer, time-resolved PIV, Tygon tubing
+Acoustic detection: microphone on valve housing detects clot formation via frequency shift
+Projects: 13 total, 3 pillars - MHV hemodynamics, CKD diagnostics, FSI simulations
 """
 
-def ask_biomni(question):
+def search_pubmed(query, n=3):
     try:
-        response = client.chat.completions.create(
-            model="llama-3.3-70b-versatile",
-            messages=[
-                {"role": "system", "content": "You are CardioLab AI built on Biomni from Stanford SNAP Lab. Expert in SJSU Biomedical Engineering research. Always give detailed accurate answers.\n" + KNOWHOW},
-                {"role": "user", "content": question}
-            ],
-            max_tokens=500
-        )
-        return response.choices[0].message.content
-    except Exception as e:
-        return "Error: " + str(e)
+        r = requests.get("https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi",
+            params={"db":"pubmed","term":query,"retmax":n,"retmode":"json"}, timeout=10)
+        ids = r.json()["esearchresult"]["idlist"]
+        if not ids: return ""
+        r2 = requests.get("https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi",
+            params={"db":"pubmed","id":",".join(ids),"retmode":"xml","rettype":"abstract"}, timeout=10)
+        import xmltodict
+        data = xmltodict.parse(r2.content)
+        articles = data.get("PubmedArticleSet",{}).get("PubmedArticle",[])
+        if isinstance(articles, dict): articles = [articles]
+        out = []
+        for a in articles[:n]:
+            try:
+                c = a["MedlineCitation"]
+                title = str(c["Article"]["ArticleTitle"])
+                abstract = c["Article"].get("Abstract",{}).get("AbstractText","")
+                if isinstance(abstract, list): abstract = " ".join([str(x) for x in abstract])
+                if isinstance(abstract, dict): abstract = str(abstract.get("#text",""))
+                pmid = str(c["PMID"]["#text"] if isinstance(c["PMID"],dict) else c["PMID"])
+                out.append("[PubMed: " + title[:80] + "]\n" + str(abstract)[:300] + "\nURL: https://pubmed.ncbi.nlm.nih.gov/" + pmid)
+            except: continue
+        return "\n\n".join(out)
+    except: return ""
+
+def search_scholar(query, n=3):
+    try:
+        r = requests.get("https://api.semanticscholar.org/graph/v1/paper/search",
+            params={"query":query,"limit":n,"fields":"title,abstract,year,url"}, timeout=10)
+        papers = r.json().get("data",[])
+        out = []
+        for p in papers:
+            out.append("[Scholar " + str(p.get("year","")) + ": " + p.get("title","")[:80] + "]\n" + (p.get("abstract") or "")[:300] + "\nURL: " + p.get("url",""))
+        return "\n\n".join(out)
+    except: return ""
+
+def ask_agent(question):
+    if not GROQ_KEY:
+        return "Error: GROQ_API_KEY not set in Space secrets."
+    cardio_query = question + " AND (mechanical heart valve OR microfluidic OR CKD creatinine OR PIV hemodynamics OR thrombogenicity)"
+    pubmed = search_pubmed(cardio_query, n=3)
+    scholar = search_scholar(question + " biomedical", n=3)
+    sources = pubmed + "\n\n" + scholar
+    response = client.chat.completions.create(
+        model="llama-3.3-70b-versatile",
+        messages=[
+            {"role":"system","content":"You are CardioLab AI built on Biomni from Stanford SNAP Lab. Expert in SJSU Biomedical Engineering. Always cite sources.\n\nCARDIOLAB KNOW-HOW:\n" + KNOWHOW},
+            {"role":"user","content":"Research question: " + question + "\n\nOnline sources found:\n" + sources[:3000]}
+        ],
+        max_tokens=800
+    )
+    return response.choices[0].message.content
 
 def piv_tool(velocity, shear, hr):
-    v = "HIGH - stenosis risk" if float(velocity) > 2.0 else "NORMAL"
-    s = "HIGH - thrombosis risk" if float(shear) > 10 else "ELEVATED - monitor" if float(shear) > 5 else "NORMAL"
-    return "Velocity: " + str(velocity) + " m/s - " + v + "\nShear Stress: " + str(shear) + " Pa - " + s + "\nHeart Rate: " + str(hr) + " bpm"
+    v = "HIGH - stenosis risk" if float(velocity)>2.0 else "NORMAL"
+    s = "HIGH - thrombosis risk" if float(shear)>10 else "ELEVATED - monitor" if float(shear)>5 else "NORMAL"
+    return "Velocity: "+str(velocity)+" m/s - "+v+"\nShear: "+str(shear)+" Pa - "+s+"\nHeart Rate: "+str(hr)+" bpm"
 
 def tgt_tool(tat, pf12, hemo, platelets, time):
     risk = sum([float(tat)>15, float(pf12)>2.0, float(hemo)>50, float(platelets)<150])
     overall = "HIGH THROMBOGENIC RISK" if risk>=3 else "MODERATE RISK" if risk>=2 else "LOW RISK"
-    return "TAT: " + str(tat) + "\nPF1.2: " + str(pf12) + "\nFree Hemoglobin: " + str(hemo) + "\nPlatelets: " + str(platelets) + "\nTime: " + str(time) + " min\nOverall: " + overall
+    return "TAT: "+str(tat)+"\nPF1.2: "+str(pf12)+"\nHemoglobin: "+str(hemo)+"\nPlatelets: "+str(platelets)+"\nTime: "+str(time)+" min\nResult: "+overall
 
 def upad_tool(r, g, b):
     creatinine = max(0, round(0.02*(float(r)-float(b))-0.5, 2))
     stage = "Normal" if creatinine<1.2 else "Borderline" if creatinine<1.5 else "Stage 2 CKD" if creatinine<3.0 else "Stage 3-4 CKD" if creatinine<6.0 else "Stage 5 CKD"
-    return "Creatinine: " + str(creatinine) + " mg/dL\nCKD Stage: " + stage + "\nConfirm with: Heska Element HT5"
+    return "Creatinine: "+str(creatinine)+" mg/dL\nCKD Stage: "+stage+"\nConfirm with: Heska Element HT5"
 
 with gr.Blocks(title="CardioLab AI - SJSU") as demo:
     gr.Markdown("# CardioLab AI Agent")
-    gr.Markdown("### SJSU Biomedical Engineering | Built on Biomni (Stanford) | Powered by Llama 70B")
-    gr.Markdown("GitHub: github.com/pranatechsol/Cardio-Lab-Ai")
-    with gr.Tab("Ask Agent"):
-        gr.Markdown("Ask anything about MCL, PIV, TGT, uPAD, CKD, FSI, MHV")
-        q = gr.Textbox(label="Your question", placeholder="e.g. What is Sylgard 184 used for?")
-        a = gr.Textbox(label="Answer", lines=6)
-        gr.Button("Ask CardioLab AI").click(ask_biomni, inputs=q, outputs=a)
+    gr.Markdown("### SJSU Biomedical Engineering | Built on Biomni (Stanford) | Llama 70B + PubMed + Semantic Scholar")
+    gr.Markdown("**Open Source** | GitHub: github.com/pranatechsol/Cardio-Lab-Ai")
+    with gr.Tab("Research Assistant"):
+        gr.Markdown("### Ask anything — searches CardioLab papers + PubMed + Semantic Scholar")
+        q = gr.Textbox(label="Research question", placeholder="e.g. What are latest methods for MHV thrombogenicity detection?")
+        a = gr.Textbox(label="Answer with citations", lines=10)
+        gr.Button("Search & Answer").click(ask_agent, inputs=q, outputs=a)
     with gr.Tab("PIV Analysis"):
         gr.Markdown("### Analyze PIV flow data from Mock Circulatory Loop")
         v = gr.Number(label="Max Velocity m/s", value=1.8)