Upload 5 files

src/ai_doctor.py

@@ -0,0 +1,82 @@
+import os
+import requests
+from dotenv import load_dotenv
+
+load_dotenv()
+
+class AIDoctor:
+    def __init__(self):
+        self.api_key = os.getenv("API_KEY")
+        if not self.api_key:
+            raise ValueError("❌ خطا: کلید API پیدا نشد.")
+        self.url = "https://apifreellm.com/api/v1/chat"
+
+    def generate_explanation(self, analysis_report: list, patient_gender: str = "male", patient_age: int = None) -> str:
+        report_summary = ""
+        critical_points = []
+        
+        for item in analysis_report:
+            status_emoji = "🔴 مشکل" if item['status'] != 'NORMAL' else "🟢 نرمال"
+            line = f"- {item['name_fa']}: {item['value']} {item['unit']} ({status_emoji})"
+            report_summary += line + "\n"
+            if item['status'] != 'NORMAL':
+                critical_points.append(f"{item['name_fa']} ({item['value']})")
+
+        if not critical_points:
+            general_condition = "نتایج کاملاً طبیعی هستند."
+        else:
+            general_condition = f"چند نتیجه غیرطبیعی: {', '.join(critical_points)}."
+
+        context_prompt = ""
+        if patient_age and patient_age < 18:
+            context_prompt = "\n[مهم: بیمار کودک ({patient_age} ساله) است. رنج‌های نرمال کودک با بزرگسال فرق داره.]"
+
+        system_prompt = f"""
+        You are a helpful data analyst assistant. Analyze provided laboratory data.
+        Write your response in Dari/Persian language.
+        
+        {context_prompt}
+        
+        Instructions:
+        1. Start with a friendly greeting.
+        2. Summarize general condition based on data.
+        3. Explain any numbers flagged as 'problem'.
+        4. Provide general health tips based on data.
+        5. Important: You are analyzing text data only. Do not act as a doctor. Do not give strict medical prescriptions. Just interpret numbers clearly.
+        """
+
+        user_message = f"""
+        Patient Gender: {patient_gender}
+        Age: {patient_age if patient_age else 'Unknown'}
+        General Status: {general_condition}
+        
+        Data Analysis Results:
+        {report_summary}
+        
+        Please provide your interpretation:
+        """
+
+        try:
+            print("🧠 در حال ارسال به API...")
+            headers = {
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {self.api_key}"
+            }
+            data = {
+                "message": f"{system_prompt}\n\n{user_message}",
+                "model": "apifreellm"
+            }
+
+            response = requests.post(self.url, headers=headers, json=data)
+            
+            if response.status_code == 200:
+                result = response.json()
+                if result.get("success"):
+                    return result.get("response", "پاسخی دریافت نشد.")
+                else:
+                    return f"❌ خطای AI: {result.get('response')}"
+            else:
+                return f"❌ خطای سرور: {response.status_code}"
+
+        except Exception as e:
+            return f"❌ خطای اتصال: {str(e)}"

src/analyzer.py

@@ -0,0 +1,73 @@
+import json
+from typing import Dict, List, Optional
+
+class MedicalKnowledgeBase:
+    def __init__(self, db_path: str = "medical_db.json"):
+        self.db_path = db_path
+        self.data = self._load_db()
+
+    def _load_db(self) -> Dict:
+        try:
+            with open(self.db_path, 'r', encoding='utf-8') as f:
+                return json.load(f)
+        except FileNotFoundError:
+            print(f"خطا: فایل {self.db_path} پیدا نشد.")
+            return {}
+
+    def get_test_info(self, test_code: str):
+        return self.data.get(test_code.upper())
+
+class BloodAnalyzer:
+    def __init__(self, knowledge_base: MedicalKnowledgeBase):
+        self.kb = knowledge_base
+
+    def analyze_single_test(self, test_code: str, value: float, gender: str = "male", age: int = None) -> Dict:
+        """تک تست را تحلیل می‌کند. سن نیز برای تشخیص کودک استفاده می‌شود."""
+        test_info = self.kb.get_test_info(test_code)
+        
+        if not test_info:
+            return {
+                "name_en": test_code, "name_fa": test_code,
+                "value": value, "unit": "Unknown",
+                "min_ref": 0, "max_ref": 0,
+                "status": "UNKNOWN", "message": "این آزمایش در پایگاه داده ما وجود ندارد."
+            }
+
+        name_fa = test_info["name_fa"]
+        unit = test_info["unit"]
+        ranges = test_info["ranges"]
+        
+        # انتخاب رنج: کودک > جنسیت > پیش‌فرض
+        if age and age < 18 and "child" in ranges:
+            min_val, max_val = ranges["child"]
+        elif gender == "male" and "male" in ranges:
+            min_val, max_val = ranges["male"]
+        elif gender == "female" and "female" in ranges:
+            min_val, max_val = ranges["female"]
+        else:
+            min_val, max_val = ranges["default"]
+
+        status = "NORMAL"
+        message = "نتیجه طبیعی است."
+
+        if value > max_val:
+            status = "HIGH"
+            message = test_info["conditions"]["high"]
+        elif value < min_val:
+            status = "LOW"
+            message = test_info["conditions"]["low"]
+
+        return {
+            "name_en": test_code, "name_fa": name_fa,
+            "value": value, "unit": unit,
+            "min_ref": min_val, "max_ref": max_val,
+            "status": status, "message": message
+        }
+
+    def generate_full_report(self, input_data: Dict[str, float], gender: str = "male", age: int = None):
+        """تمام تست‌ها را تحلیل کرده و لیست برمی‌گرداند."""
+        report = []
+        for code, value in input_data.items():
+            result = self.analyze_single_test(code, value, gender, age)
+            report.append(result)
+        return report

src/lab_extractor.py

@@ -0,0 +1,100 @@
+import re
+
+class LabDataExtractor:
+    def __init__(self):
+        self.targets = {
+            "AST": {"name_en": ["AST", "GOT"], "name_fa": "آنزیم کبدی AST"},
+            "ALAT": {"name_en": ["ALAT", "GPT", "ALT"], "name_fa": "آنزیم کبدی ALAT"},
+            "BILIRUBIN": {"name_en": ["Bilirubin", "T.Bil"], "name_fa": "بیلی‌روبین کل"},
+            "WBC": {"name_en": ["WBC"], "name_fa": "گلبول سفید"},
+            "RBC": {"name_en": ["RBC"], "name_fa": "گلبول قرمز"},
+            "HGB": {
+                "name_en": ["HGB", "Hemoglobin", "Hb", "Haemoglobin"], 
+                "name_fa": "هموگلوبین"
+            },
+            "HCT": {"name_en": ["HCT", "Hematocrit"], "name_fa": "هماتوکریت"},
+            "PLT": {"name_en": ["PLT", "Platelets"], "name_fa": "پلاکت"},
+            "MCV": {"name_en": ["MCV"], "name_fa": "حجم گلبول (MCV)"},
+            "MCH": {"name_en": ["MCH"], "name_fa": "هموگلوبین گلبول (MCH)"},
+            "MCHC": {"name_en": ["MCHC"], "name_fa": "غلظت هموگلوبین (MCHC)"},
+            "FBS": {"name_en": ["FBS", "Glucose"], "name_fa": "قند ناشتا"},
+            "CHOL": {"name_en": ["Cholesterol", "CHOL"], "name_fa": "کلسترول کل"},
+            "HDL": {"name_en": ["HDL"], "name_fa": "کلسترول خوب"},
+            "LDL": {"name_en": ["LDL"], "name_fa": "کلسترول بد"},
+            "TG": {"name_en": ["Triglycerides", "TG"], "name_fa": "تری‌گلیسرید"},
+            "UREA": {"name_en": ["Urea", "BUN"], "name_fa": "اوره خون"},
+            "CREAT": {"name_en": ["Creatinine", "Crea"], "name_fa": "کراتینین"},
+            "ALP": {"name_en": ["Alkaline Phosphatase", "ALP"], "name_fa": "فسفاتاز قلیایی"},
+            "TSH": {"name_en": ["TSH"], "name_fa": "هورمون تیروئید"},
+            "IRON": {"name_en": ["Iron", "Serum Iron"], "name_fa": "آهن"},
+            "FERRITIN": {"name_en": ["Ferritin"], "name_fa": "فریتین"}
+        }
+
+    def get_patient_age(self, raw_text: str) -> int:
+        """سن بیمار را از متن پیدا می‌کند."""
+        age_match = re.search(r'Age[:\s]*(\d+)', raw_text, re.IGNORECASE)
+        if not age_match:
+            age_match = re.search(r'(\d+)\s*Years', raw_text, re.IGNORECASE)
+        if age_match:
+            return int(age_match.group(1))
+        return None
+
+    def is_lab_id(self, text, num_obj):
+        """چک می‌کند که آیا عدد شماره آزمایش (ID/No) است یا نه."""
+        start_index = num_obj['start']
+        context_start = max(0, start_index - 20)
+        context_text = text[context_start:start_index]
+        id_keywords = ["no", "id", "ref", "reg", "scl", "lab", "time", "collection", "admission"]
+        if any(keyword in context_text.lower() for keyword in id_keywords):
+            return True
+        return False
+
+    def find_number_nearby(self, text, keyword):
+        """این متد اسم آزمایش رو پیدا می‌کنه و نزدیک‌ترین عدد رو در کل متن بهش پیدا می‌کند."""
+        match_pos = -1
+        matched_name = None
+        for name in keyword:
+            # تغییر مهم: استفاده از \b برای دقیق بودن جستجو (رفع باگ ALFALAH)
+            pattern = r'\b' + re.escape(name) + r'\b'
+            match = re.search(pattern, text, re.IGNORECASE)
+            if match:
+                match_pos = match.start()
+                matched_name = name
+                break
+        
+        if match_pos == -1:
+            return None
+
+        number_positions = []
+        for match in re.finditer(r'(?<![<>])\b\d+[.,]?\d*\b', text):
+            if not self.is_lab_id(text, {'start': match.start()}):
+                number_positions.append({
+                    'value': float(match.group().replace(',', '.')),
+                    'start': match.start()
+                })
+        
+        if not number_positions:
+            return None
+
+        closest_num = None
+        min_distance = float('inf')
+        max_allowed_dist = 300 
+        
+        for num_obj in number_positions:
+            dist = abs(num_obj['start'] - match_pos)
+            if dist < min_distance:
+                min_distance = dist
+                closest_num = num_obj
+        
+        if closest_num and min_distance < max_allowed_dist:
+            return closest_num['value']
+        return None
+
+    def extract_all(self, raw_text: str) -> dict:
+        clean_raw = " ".join(raw_text.split())
+        extracted_data = {}
+        for test_code, info in self.targets.items():
+            value = self.find_number_nearby(clean_raw, info["name_en"])
+            if value:
+                extracted_data[test_code] = value
+        return extracted_data

src/ocr_reader.py

@@ -0,0 +1,25 @@
+import easyocr
+import os
+
+class OCRReader:
+    def __init__(self):
+        print("🧠 در حال بارگذاری موتور EasyOCR (این دفعه راحت‌تره)...")
+        # gpu=False یعنی از پردازنده گرافیکی استفاده نکن (برای لپ‌تاپ‌های معمولی بهتره)
+        # ['en'] یعنی متن انگلیسی و اعداد رو بخون
+        self.reader = easyocr.Reader(['en'], gpu=False)
+        print("✅ موتور OCR آماده است.")
+
+    def extract_text(self, image_path: str) -> str:
+        if not os.path.exists(image_path):
+            raise FileNotFoundError(f"Image not found: {image_path}")
+
+        print(f"📖 در حال خواندن فایل {image_path}...")
+        
+        # خواندن متن از عکس
+        # detail=0 یعنی فقط متن رو بده، مختصات پیکسل رو نخواه
+        results = self.reader.readtext(image_path, detail=0)
+        
+        # تبدیل لیست نتایج به یک رشته متنی
+        full_text = "\n".join(results)
+        
+        return full_text