Upload 16 files

README.md

@@ -1,14 +1,89 @@
----
-title: Microbial Susceptibility Analyzer
-emoji: 🐢
-colorFrom: gray
-colorTo: pink
-sdk: streamlit
-sdk_version: 1.43.0
-app_file: app.py
-pinned: false
-license: mit
-short_description: Welcome to the Microbial Susceptibility Analyzer! This app h
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Microbial Susceptibility Analyzer
+
+## Overview
+The Microbial Susceptibility Analyzer is a Streamlit-based web application that combines machine learning and rule-based decision making to predict antibiotic resistance patterns. The application provides:
+- Susceptibility predictions for organism-antibiotic combinations
+- Rule-based guidance for treatment decisions
+- AI-powered explanations and recommendations
+- Batch prediction capabilities for CSV datasets
+
+## Installation
+
+1. Clone the repository:
+```bash
+git clone https://github.com/yourusername/microbial-susceptibility-analyzer.git
+cd microbial-susceptibility-analyzer
+```
+
+2. Create a virtual environment:
+```bash
+python -m venv venv
+source venv/bin/activate  # On Windows use: venv\Scripts\activate
+```
+
+3. Install dependencies:
+```bash
+pip install -r requirements.txt
+```
+
+4. Set up environment variables:
+Create a `.streamlit/secrets.toml` file with your API keys:
+```toml
+OPENAI_API_KEY = "your_openai_api_key"
+OPENAI_BASE_URL = "https://api.openai.com/v1"
+```
+
+## Usage
+
+Run the application:
+```bash
+streamlit run app.py
+```
+
+The application provides several pages:
+- **Home**: Project overview and introduction
+- **Susceptibility Analysis**: Single prediction interface
+- **AI Assistant**: Expert system for detailed explanations
+- **Data Upload**: Batch prediction for CSV files
+- **About**: Project information and contact details
+
+## File Structure
+
+```
+microbial-susceptibility-analyzer/
+├── app.py                 # Main application file
+├── prediction.py          # Prediction logic
+├── ai_assistant.py        # AI integration
+├── utils.py               # Utility functions
+├── config.py              # Configuration settings
+├── requirements.txt       # Python dependencies
+├── README.md              # This documentation
+├── label_encoders/        # Label encoding files
+│   ├── organism_label_encoder.json
+│   ├── antibiotic_label_encoder.json
+│   └── ...
+├── models/                # Trained models
+│   ├── best_model.pkl
+│   └── best_model_1.pkl
+└── data/                  # Data files
+    ├── implied_susceptibility_rules.csv
+    └── microbiology_cultures_implied_susceptibility.csv
+```
+
+## Configuration
+
+The application requires the following configuration:
+
+1. **Model Path**: Set in `config.py`
+2. **Encoder Directory**: Set in `config.py`
+3. **OpenAI API**: Configure in `.streamlit/secrets.toml`
+
+## License
+
+This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+
+## Contact
+
+For questions or support, contact:  
+Chukwuebuka Anulunko  
+[email protected]

ai_assistant.py

@@ -0,0 +1,39 @@
+# ai_assistant.py
+
+from openai import OpenAI
+import re
+
+def initialize_openai_client(api_key, base_url):
+    """Initialize the OpenAI client."""
+    return OpenAI(api_key=api_key, base_url=base_url)
+
+def generate_system_message(current_organism, current_antibiotic, current_was_positive):
+    """Generate a dynamic system message for the AI Assistant."""
+    system_message = """
+    You are a clinical microbiology AI assistant. Analyze susceptibility results considering:
+    1) Model predictions 2) CLSI/EUCAST guidelines 3) Local resistance patterns.
+    Provide: Interpretation, therapy options, resistance mechanisms, and infection control measures.
+    """
+    
+    if all([current_organism, current_antibiotic, current_was_positive is not None]):
+        system_message += f"""
+        CONTEXTUAL NOTES:
+        - Current organism: {current_organism}
+        - Current antibiotic: {current_antibiotic}
+        - Culture positivity: {'Positive' if current_was_positive else 'Negative'}
+        """
+    
+    return system_message
+
+# In ai_assistant.py
+
+def get_ai_response(client, prompt):
+    """Get a response from the AI Assistant."""
+    response = client.chat.completions.create(
+        model="deepseek-chat",
+        messages=[
+            {"role": "system", "content": "You are a clinical microbiology AI assistant."},
+            {"role": "user", "content": prompt}
+        ]
+    ).choices[0].message.content
+    return response

best_model_1.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02eeb4beb89219d8a0afdf20a5a2cc9c1ea6e77937214568cd4287f8c815eee4
+size 15174394

config.py

@@ -0,0 +1,10 @@
+# config.py
+
+
+# File paths
+MODEL_PATH = 'best_model_1.pkl'
+ENCODER_DIR = 'label_encoders'
+
+# OpenAI API key (stored in Streamlit secrets)
+OPENAI_API_KEY =st.secrets["DEEPSEEK_API_KEY"]
+OPENAI_BASE_URL = "https://api.deepseek.com"

implied_susceptibility_rules.csv

@@ -0,0 +1,101 @@
+organism,antibiotic,implied_susceptibility,rationale
+ACINETOBACTER,Aztreonam,Resistant,
+ACINETOBACTER,Cefazolin,Resistant,
+ACINETOBACTER,Minocycline,Resistant,
+ACINETOBACTER,Tetracycline,Resistant,
+ACINETOBACTER,Doripenem,Same as Meropenem if reported,
+ACINETOBACTER,Meropenem,Susceptible if susceptible to Imipenem,
+CITROBACTER,Ampicillin,Resistant,
+CITROBACTER,Cefazolin,Resistant,
+CITROBACTER,Cefotetan,Resistant,
+CITROBACTER,Cefoxitin,Resistant,
+CITROBACTER,Cefepime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+CITROBACTER,Ceftazidime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+ACINETOBACTER,Ceftriaxone,Same as Cefotaxime if reported,
+ACINETOBACTER,Cefotaxime,Same as Ceftriaxone if reported,
+ACINETOBACTER,Ertapenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+ACINETOBACTER,Imipenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+ACINETOBACTER,Meropenem,Susceptible if susceptible to Imipenem, Ceftriaxone, or Cefotaxime,
+Enterobacter,Ampicillin,Resistant,
+Enterobacter,Cefazolin,Resistant,
+Enterobacter,Cefepime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Ceftazidime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Ceftriaxone,Same as Cefotaxime if reported,
+Enterobacter,Cefotaxime,Same as Ceftriaxone if reported,
+Enterobacter,Doxycycline,Susceptible if susceptible to Tetracycline, otherwise Resistant,
+Enterobacter,Doripenem,Same as Meropenem if reported,
+Enterobacter,Ertapenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Imipenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Meropenem,Susceptible if susceptible to Imipenem, Ceftriaxone, or Cefotaxime,
+ESCHERICHIA COLI,Cefepime,Susceptible if susceptible to Ceftriaxone, Cefotaxime, or Cefazolin,
+ESCHERICHIA COLI,Ceftazidime,Susceptible if susceptible to Ceftriaxone, Cefotaxime, or Cefazolin,
+ESCHERICHIA COLI,Ceftriaxone,Same as Cefotaxime if reported,
+ESCHERICHIA COLI,Cefotetan,Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Cefoxitin,Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Cefpodoxime,Same as Ceftriaxone if reported; otherwise, Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Cefuroxime,Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Doxycycline,Susceptible if susceptible to Tetracycline, otherwise Resistant,
+KLEBSIELLA,Ampicillin,Resistant,
+Proteus species,Tetracycline,Resistant,
+Proteus species,Tigecycline,Resistant,
+Proteus species,Colistin,Resistant,
+Pseudomonas aeruginosa,Ampicillin,Resistant,
+Pseudomonas aeruginosa,Ceftriaxone,Resistant,
+Pseudomonas aeruginosa,Cefazolin,Resistant,
+Pseudomonas aeruginosa,Ertapenem,Resistant,
+Pseudomonas aeruginosa,Tetracycline,Resistant,
+Pseudomonas aeruginosa,Tigecycline,Resistant,
+Serratia,Ampicillin,Resistant,
+Serratia,Cefazolin,Resistant,
+Serratia,Tetracycline,Resistant
+organism,antibiotic,implied_susceptibility,rationale
+ACINETOBACTER,Aztreonam,Resistant,
+ACINETOBACTER,Cefazolin,Resistant,
+ACINETOBACTER,Minocycline,Resistant,
+ACINETOBACTER,Tetracycline,Resistant,
+ACINETOBACTER,Doripenem,Same as Meropenem if reported,
+ACINETOBACTER,Meropenem,Susceptible if susceptible to Imipenem,
+CITROBACTER,Ampicillin,Resistant,
+CITROBACTER,Cefazolin,Resistant,
+CITROBACTER,Cefotetan,Resistant,
+CITROBACTER,Cefoxitin,Resistant,
+CITROBACTER,Cefepime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+CITROBACTER,Ceftazidime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+ACINETOBACTER,Ceftriaxone,Same as Cefotaxime if reported,
+ACINETOBACTER,Cefotaxime,Same as Ceftriaxone if reported,
+ACINETOBACTER,Ertapenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+ACINETOBACTER,Imipenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+ACINETOBACTER,Meropenem,Susceptible if susceptible to Imipenem, Ceftriaxone, or Cefotaxime,
+Enterobacter,Ampicillin,Resistant,
+Enterobacter,Cefazolin,Resistant,
+Enterobacter,Cefepime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Ceftazidime,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Ceftriaxone,Same as Cefotaxime if reported,
+Enterobacter,Cefotaxime,Same as Ceftriaxone if reported,
+Enterobacter,Doxycycline,Susceptible if susceptible to Tetracycline, otherwise Resistant,
+Enterobacter,Doripenem,Same as Meropenem if reported,
+Enterobacter,Ertapenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Imipenem,Susceptible if susceptible to Ceftriaxone or Cefotaxime,
+Enterobacter,Meropenem,Susceptible if susceptible to Imipenem, Ceftriaxone, or Cefotaxime,
+ESCHERICHIA COLI,Cefepime,Susceptible if susceptible to Ceftriaxone, Cefotaxime, or Cefazolin,
+ESCHERICHIA COLI,Ceftazidime,Susceptible if susceptible to Ceftriaxone, Cefotaxime, or Cefazolin,
+ESCHERICHIA COLI,Ceftriaxone,Same as Cefotaxime if reported,
+ESCHERICHIA COLI,Cefotetan,Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Cefoxitin,Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Cefpodoxime,Same as Ceftriaxone if reported; otherwise, Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Cefuroxime,Susceptible if susceptible to Cefazolin,
+ESCHERICHIA COLI,Doxycycline,Susceptible if susceptible to Tetracycline, otherwise Resistant,
+KLEBSIELLA,Ampicillin,Resistant,
+Proteus species,Tetracycline,Resistant,
+Proteus species,Tigecycline,Resistant,
+Proteus species,Colistin,Resistant,
+Pseudomonas aeruginosa,Ampicillin,Resistant,
+Pseudomonas aeruginosa,Ceftriaxone,Resistant,
+Pseudomonas aeruginosa,Cefazolin,Resistant,
+Pseudomonas aeruginosa,Ertapenem,Resistant,
+Pseudomonas aeruginosa,Tetracycline,Resistant,
+Pseudomonas aeruginosa,Tigecycline,Resistant,
+Serratia,Ampicillin,Resistant,
+Serratia,Cefazolin,Resistant,
+Serratia,Cefazolin,Resistant
+Serratia,Tetracycline,Resistant

label_encoders/antibiotic_label_encoder.json

@@ -0,0 +1 @@
+{"Amikacin": 0, "Amoxicillin/Clavulanic Acid": 1, "Ampicillin": 2, "Ampicillin/Sulbactam": 3, "Aztreonam": 4, "Cefazolin": 5, "Cefepime": 6, "Cefiderocol": 7, "Cefotaxime": 8, "Cefotetan": 9, "Cefoxitin": 10, "Cefpodoxime": 11, "Ceftaroline": 12, "Ceftazidime": 13, "Ceftazidime/Avibactam": 14, "Ceftolozane/Tazobactam": 15, "Ceftriaxone": 16, "Cefuroxime": 17, "Cephalexin/Cephalothin": 18, "Chloramphenicol": 19, "Ciprofloxacin": 20, "Clarithromycin": 21, "Clindamycin": 22, "Colistin": 23, "Daptomycin": 24, "Doripenem": 25, "Doxycycline": 26, "Ertapenem": 27, "Erythromycin": 28, "Fosfomycin": 29, "Gentamicin": 30, "Imipenem": 31, "Imipenem/Ebactam": 32, "Levofloxacin": 33, "Linezolid": 34, "Meropenem": 35, "Meropenem/Vaborbactam": 36, "Metronidazole": 37, "Minocycline": 38, "Moxifloxacin": 39, "Nitrofurantoin": 40, "Null": 41, "Oxacillin": 42, "Penicillin": 43, "Piperacillin": 44, "Piperacillin/Tazobactam": 45, "Quinupristin/Dalfopristin": 46, "Streptomycin": 47, "Tetracycline": 48, "Ticacarcillin/Clavulanic Acid": 49, "Ticarcillin": 50, "Tigecycline": 51, "Tobramycin": 52, "Trimethoprim/Sulfamethoxazole": 53, "Vancomycin": 54}

label_encoders/culture_description_label_encoder.json

@@ -0,0 +1 @@
+{"BLOOD": 0, "RESPIRATORY": 1, "URINE": 2}

label_encoders/ordering_mode_label_encoder.json

@@ -0,0 +1 @@
+{"Inpatient": 0, "Null": 1, "Outpatient": 2}

label_encoders/organism_label_encoder.json

@@ -0,0 +1 @@
+{"ABIOTROPHIA SPECIES (OR GRANULICATELLA SPECIES)": 0, "ACHROMOBACTER SPECIES": 1, "ACHROMOBACTER XYLOSOXIDANS": 2, "ACID FAST BACILLI": 3, "ACID FAST BACILLI DETECTED IN LIQUID MEDIUM": 4, "ACINETOBACTER BAUMANNII": 5, "ACINETOBACTER BAUMANNII COMPLEX": 6, "ACINETOBACTER HAEMOLYTICUS": 7, "ACINETOBACTER LWOFFI": 8, "ACINETOBACTER SPECIES": 9, "ACTINOMYCES SPECIES": 10, "AEROCOCCUS SANGUINICOLA": 11, "AEROCOCCUS SPECIES": 12, "AEROCOCCUS URINAE": 13, "AEROMONAS CAVIAE": 14, "AEROMONAS HYDROPHILA GROUP": 15, "AEROMONAS SPECIES": 16, "AEROMONAS VERONII": 17, "ALCALIGENES FAECALIS": 18, "ALCALIGENES SPECIES": 19, "ANAEROBIC GRAM NEGATIVE RODS": 20, "ANAEROBIC GRAM POSITIVE COCCI": 21, "ANAEROBIC GRAM POSITIVE RODS": 22, "ARCANOBACTERIUM HAEMOLYTICUM": 23, "BACILLUS SPECIES": 24, "BACTEROIDES CACCAE": 25, "BACTEROIDES FRAGILIS": 26, "BACTEROIDES OVATUS": 27, "BACTEROIDES SPECIES": 28, "BACTEROIDES SPECIES NOT FRAGILIS": 29, "BACTEROIDES THETAIOTAOMICRON": 30, "BACTEROIDES UNIFORMIS": 31, "BACTEROIDES UREOLYTICUS": 32, "BACTEROIDES VULGATUS": 33, "BETA STREPTOCOCCUS GROUP C": 34, "BETA STREPTOCOCCUS GROUP G": 35, "BETA STREPTOCOCCUS SPECIES": 36, "BORDETELLA BRONCHISEPTICA": 37, "BREVIBACTERIUM SPECIES": 38, "BREVUNDIMONAS": 39, "BURKHOLDERIA CENOCEPACIA": 40, "BURKHOLDERIA CEPACIA": 41, "BURKHOLDERIA CEPACIA COMPLEX": 42, "BURKHOLDERIA CEPACIA COMPLEX GENOMOVAR I": 43, "BURKHOLDERIA GLADIOLI": 44, "BURKHOLDERIA MULTIVORANS": 45, "BURKHOLDERIA VIETNAMIENSIS": 46, "CAMPYLOBACTER SPECIES": 47, "CHROMOGENIC AFB, NOT M.TUBERCULOSIS": 48, "CHRYSEOBACTERIUM INDOLOGENES": 49, "CHRYSEOBACTERIUM SPECIES": 50, "CHRYSEOMONAS LUTEOLA": 51, "CITROBACTER AMALONATICUS COMPLEX": 52, "CITROBACTER BRAAKII": 53, "CITROBACTER FREUNDII": 54, "CITROBACTER FREUNDII COMPLEX": 55, "CITROBACTER FREUNDII COMPLEX (CARBAPENEM RESISTANT)": 56, "CITROBACTER KOSERI": 57, "CITROBACTER KOSERI (CARBAPENEM RESISTANT)": 58, "CITROBACTER SPECIES": 59, "CITROBACTER YOUNGAE": 60, "CLOSTRIDIUM CLOSTRIDIOFORME": 61, "CLOSTRIDIUM INNOCUUM": 62, "CLOSTRIDIUM PARAPUTRIFICUM": 63, "CLOSTRIDIUM PERFRINGENS": 64, "CLOSTRIDIUM RAMOSUM": 65, "CLOSTRIDIUM SEPTICUM": 66, "CLOSTRIDIUM SORDELLII": 67, "CLOSTRIDIUM SPECIES": 68, "CLOSTRIDIUM SPECIES, NOT PERFRINGENS": 69, "CLOSTRIDIUM TERTIUM": 70, "COAG NEGATIVE STAPHYLOCOCCUS": 71, "COMAMONAS ACIDOVORANS": 72, "CORYNEBAC. STRIATUM/AMYCOLATUM": 73, "CORYNEBACTERIUM ACCOLENS": 74, "CORYNEBACTERIUM AMYCOLATUM": 75, "CORYNEBACTERIUM DIPTHERIAE": 76, "CORYNEBACTERIUM GROUP F-1": 77, "CORYNEBACTERIUM JEIKEIUM": 78, "CORYNEBACTERIUM MACGINLEYI": 79, "CORYNEBACTERIUM PROPINQUUM": 80, "CORYNEBACTERIUM PSEUDODIPHTHERITICUM": 81, "CORYNEBACTERIUM SPECIES": 82, "CORYNEBACTERIUM SPECIES, NOT C. UREALYTICUM": 83, "CORYNEBACTERIUM STRIATUM": 84, "CORYNEBACTERIUM UREALYTICUM": 85, "CRONOBACTER SAKAZAKII": 86, "CULTURE POSITIVE GRAM NEGATIVE COCCI": 87, "CULTURE POSITIVE GRAM POSITIVE COCCI": 88, "DERMABACTER HOMINIS": 89, "DIPHTHEROIDS - CORYNEBACTERIUM SPECIES": 90, "EGGERTHELLA LENTA": 91, "ELIZABETHKINGIA MENINGOSEPTICA": 92, "ELIZABETHKINGIA SPECIES": 93, "ENTERIC GRAM NEGATIVE RODS": 94, "ENTEROBACTER AMNIGENUS 1": 95, "ENTEROBACTER ASBURIAE": 96, "ENTEROBACTER CANCEROGENUS": 97, "ENTEROBACTER CLOACAE": 98, "ENTEROBACTER CLOACAE COMPLEX": 99, "ENTEROBACTER CLOACAE COMPLEX (CARBAPENEM RESISTANT)": 100, "ENTEROBACTER GERGOVIAE": 101, "ENTEROBACTER SPECIES": 102, "ENTEROBACTER TAYLORAE": 103, "ENTEROCOCCUS AVIUM": 104, "ENTEROCOCCUS CASSELIFLAVUS": 105, "ENTEROCOCCUS DURANS": 106, "ENTEROCOCCUS DURANS/HIRAE": 107, "ENTEROCOCCUS FAECALIS": 108, "ENTEROCOCCUS FAECALIS - VANCO RESISTANT": 109, "ENTEROCOCCUS FAECIUM": 110, "ENTEROCOCCUS FAECIUM -  VANCO RESISTANT": 111, "ENTEROCOCCUS GALLINARUM": 112, "ENTEROCOCCUS RAFFINOSUS": 113, "ENTEROCOCCUS SPECIES": 114, "ESCHERICHIA COLI": 115, "ESCHERICHIA COLI (CARBAPENEM RESISTANT)": 116, "ESCHERICHIA FERGUSONII": 117, "ESCHERICHIA HERMANII": 118, "ESCHERICHIA SPECIES": 119, "ESCHERICHIA VULNERIS": 120, "EUBACTERIUM LENTUM": 121, "EWINGELLA AMERICANA": 122, "FLAVOBACTERIUM SPECIES": 123, "FUSOBACTERIUM NECROPHORUM": 124, "FUSOBACTERIUM NUCLEATUM": 125, "FUSOBACTERIUM SPECIES": 126, "GORDONIA SPECIES": 127, "GRAM NEGATIVE RODS": 128, "GRAM NEGATIVE RODS - NONFERM, UNIDENTIFIED": 129, "GRAM POSITIVE RODS": 130, "HAEMOPHILUS ACTINOMYCETEMCOMITANS": 131, "HAFNIA": 132, "HAFNIA ALVEI": 133, "HAFNIA ALVEI (CARBAPENEM RESISTANT)": 134, "KLEBSIELLA AEROGENES": 135, "KLEBSIELLA AEROGENES (CARBAPENEM RESISTANT)": 136, "KLEBSIELLA OXYTOCA": 137, "KLEBSIELLA OZAENAE": 138, "KLEBSIELLA PNEUMONIAE": 139, "KLEBSIELLA PNEUMONIAE (CARBAPENEM RESISTANT)": 140, "KLEBSIELLA PNEUMONIAE SSP. OZAENAE": 141, "KLEBSIELLA SPECIES": 142, "KLUYVERA ASCORBATA": 143, "KLUYVERA CRYOCRESCENS": 144, "KLUYVERA SPECIES": 145, "LACTOBACILLUS CASEI": 146, "LACTOBACILLUS RHAMNOSUS": 147, "LACTOBACILLUS SPECIES": 148, "LACTOCOCCUS SPECIES": 149, "LACTOSE FERMENTING GRAM NEGATIVE RODS": 150, "LECLERCIA ADECARBOXYLATA": 151, "LEUCONOSTOC SPECIES": 152, "LISTERIA MONOCYTOGENES": 153, "LISTERIA SPECIES": 154, "MICROBACTERIUM SPECIES": 155, "MICROCOCCUS SPECIES": 156, "MIXED GENITAL FLORA": 157, "MORAXELLA NON-LIQUEFACIENS": 158, "MORAXELLA OSLOENSIS": 159, "MORAXELLA SPECIES": 160, "MORGANELLA MORGANII": 161, "MUCOID PSEUDOMONAS AERUGINOSA": 162, "MYCOBACTERIUM ABSCESSUS": 163, "MYCOBACTERIUM ABSCESSUS GROUP": 164, "MYCOBACTERIUM AVIUM COMPLEX": 165, "MYCOBACTERIUM AVIUM COMPLEX BY DNA PROBE": 166, "MYCOBACTERIUM CHELONAE": 167, "MYCOBACTERIUM FORTUITUM": 168, "MYCOBACTERIUM FORTUITUM SPECIES GROUP": 169, "MYCOBACTERIUM IMMUNOGENUM": 170, "MYCOBACTERIUM NEOAURUM": 171, "MYCOBACTERIUM SPECIES": 172, "MYCOBACTERIUM SPECIES - RAPID GROWER": 173, "MYCOBACTERIUM TUBERCULOSIS COMPLEX BY DNA PROBE": 174, "MYCOBACTERIUM XENOPI": 175, "MYROIDES SPECIES": 176, "NOCARDIA ASTEROIDES": 177, "NOCARDIA CYRIACIGEORGICA": 178, "NOCARDIA FARCINICA": 179, "NOCARDIA NOVA": 180, "NOCARDIA NOVA COMPLEX": 181, "NOCARDIA SPECIES": 182, "NOCARDIA TRANSVALENSIS": 183, "NOCARDIA TRANSVALENSIS COMPLEX": 184, "NON FERMENTING GRAM NEGATIVE RODS": 185, "NON LACTOSE FERMENTING GRAM NEGATIVE RODS": 186, "NON SPORE-FORMING ANAEROBIC GRAM POSITIVE RODS": 187, "NON-FERMENTING GRAM NEGATIVE RODS - NOT P.AERUGINOSA": 188, "NONFERMENTING GRAM NEGATIVE RODS - UNIDENTIFIED": 189, "NORCARDIA WALLACEI": 190, "Null": 191, "OCHROBACTRUM": 192, "OCHROBACTRUM ANTHROPI": 193, "OLIGELLA URETHRALIS": 194, "PAENIBACILLUS SPECIES": 195, "PANTOEA AGGLOMERANS": 196, "PANTOEA SPECIES": 197, "PARABACTEROIDES DISTASONIS": 198, "PARABACTEROIDES SPECIES": 199, "PASTEURELLA MULTOCIDA": 200, "PEDIOCOCCUS SP.": 201, "PEPTOSTREPTOCOCCUS SPECIES": 202, "PREVOTELLA SPECIES": 203, "PROTEUS MIRABILIS": 204, "PROTEUS SPECIES": 205, "PROTEUS VULGARIS GROUP": 206, "PROVIDENCIA RETTGERI": 207, "PROVIDENCIA SPECIES": 208, "PROVIDENCIA STUARTII": 209, "PSEUDOMONAS AERUGINOSA": 210, "PSEUDOMONAS AERUGINOSA (NON-MUCOID CF)": 211, "PSEUDOMONAS FLUORESCENS": 212, "PSEUDOMONAS FLUORESCENS GROUP": 213, "PSEUDOMONAS MENDOCINA": 214, "PSEUDOMONAS ORYZIHABITANS": 215, "PSEUDOMONAS PUTIDA": 216, "PSEUDOMONAS PUTIDA GROUP": 217, "PSEUDOMONAS SPECIES": 218, "PSEUDOMONAS SPECIES - FLUORESCENS/PUTIDA GROUP": 219, "PSEUDOMONAS SPECIES - NOT P.AERUGINOSA": 220, "PSEUDOMONAS STUTZERI": 221, "PSEUDOMONAS STUTZERI GROUP": 222, "RAHNELLA": 223, "RALSTONIA MANNITOLILYTICA": 224, "RALSTONIA SPECIES": 225, "RAOULTELLA": 226, "RAOULTELLA ORNITHINOLYTICA": 227, "RAOULTELLA PLANTICOLA": 228, "RHIZOBIUM RADIOBACTER": 229, "SALMONELLA ENTERICA": 230, "SALMONELLA SPECIES": 231, "SALMONELLA TYPHI": 232, "SCHISTOSOMA HAEMATOBIUM OVA": 233, "SERRATIA LIQUEFACIENS": 234, "SERRATIA MARCESCENS": 235, "SERRATIA MARCESCENS (CARBAPENEM RESISTANT)": 236, "SERRATIA ODORIFERA": 237, "SERRATIA PLYMUTHICA": 238, "SERRATIA RUBIDAEA": 239, "SERRATIA SPECIES": 240, "SHEWANELLA PUTREFACIENS": 241, "SHIGELLA SPECIES": 242, "SPHINGOBACTERIUM MULTIVORUM": 243, "SPHINGOBACTERIUM SPIRITIVORUM": 244, "SPHINGOMONAS": 245, "SPHINGOMONAS PAUCIMOBILIS/PARAPAUCIMOBILIS": 246, "STAPH AUREUS {MRSA}": 247, "STAPH AUREUS(COLONY VARIANT - SMALL COLONY OR OTHER MORPHOTYPE)": 248, "STAPH NEG NOT LUGDUNENSIS": 249, "STAPHYLOCOCCUS AUREUS": 250, "STAPHYLOCOCCUS COHNII": 251, "STAPHYLOCOCCUS EPIDERMIDIS": 252, "STAPHYLOCOCCUS HAEMOLYTICUS": 253, "STAPHYLOCOCCUS LUGDUNENSIS": 254, "STAPHYLOCOCCUS SAPROPHYTICUS": 255, "STAPHYLOCOCCUS SPECIES": 256, "STAPHYLOCOCCUS SPECIES - COAG NEGATIVE, NOT STAPH SAPROPHYTICUS": 257, "STAPHYLOCOCCUS SPECIES -NOT STAPH. AUREUS": 258, "STAPHYLOCOCCUS SPECIES, COAG PENDING": 259, "STENOTROPHOMONAS MALTOPHILIA": 260, "STREP. GRP D NOT ENTEROCOCCI": 261, "STREPTOCOCCUS AGALACTIAE (GROUP B)": 262, "STREPTOCOCCUS AGALACTIAE {GROUP B}": 263, "STREPTOCOCCUS ANGINOSUS": 264, "STREPTOCOCCUS ANGINOSUS GROUP": 265, "STREPTOCOCCUS BOVIS": 266, "STREPTOCOCCUS BOVIS GP.": 267, "STREPTOCOCCUS CONSTELLATUS": 268, "STREPTOCOCCUS DYSGALACTIAE": 269, "STREPTOCOCCUS DYSGALACTIAE SUBSP. EQUISIMILIS (GROUP C/G STR]": 270, "STREPTOCOCCUS GALLOLYTICUS": 271, "STREPTOCOCCUS GALLOLYTICUS, SUBSP. GALLOLYTICUS": 272, "STREPTOCOCCUS GALLOLYTICUS, SUBSP. PASTEURIANUS": 273, "STREPTOCOCCUS GORDONII": 274, "STREPTOCOCCUS INTERMEDIUS": 275, "STREPTOCOCCUS MITIS": 276, "STREPTOCOCCUS MITIS GROUP": 277, "STREPTOCOCCUS MITIS/STREPTOCOCCUS ORALIS": 278, "STREPTOCOCCUS MUTANS": 279, "STREPTOCOCCUS MUTANS GROUP": 280, "STREPTOCOCCUS ORALIS": 281, "STREPTOCOCCUS PARASANGUINIS": 282, "STREPTOCOCCUS PNEUMONIAE": 283, "STREPTOCOCCUS PYOGENES (GROUP A)": 284, "STREPTOCOCCUS PYOGENES {GROUP A}": 285, "STREPTOCOCCUS SALIVARIUS": 286, "STREPTOCOCCUS SALIVARIUS GROUP": 287, "STREPTOCOCCUS SANGUINIS": 288, "STREPTOCOCCUS SANGUIS": 289, "STREPTOCOCCUS SPECIES": 290, "STREPTOCOCCUS SPECIES - ALPHA HEMOLYTIC": 291, "STREPTOCOCCUS SPECIES - ALPHA HEMOLYTIC, NOT S.PNEUMO": 292, "TSUKAMURELLA PULMONIS": 293, "TSUKAMURELLA SPECIES": 294, "VEILLONELLA SPECIES": 295, "VIRIDANS GROUP STREPTOCOCCI": 296, "VIRIDANS GROUP STREPTOCOCCI NOT S. PNEUMONIAE": 297, "YERSINIA ENTEROCOLITICA": 298, "ZZAEROMONAS CAVIAE COMPLEX": 299, "ZZAEROMONAS HYDROPHILA COMPLEX": 300, "ZZAEROMONAS VERONII COMPLEX": 301, "ZZZBACTEROIDES DISTASONIS": 302, "ZZZBACTEROIDES FRAGILIS GROUP": 303, "ZZZBACTEROIDES NOT B. FRAGILIS GRP": 304, "ZZZCITROBACTER AMALONATICUS": 305, "ZZZCITROBACTER FARMERI": 306, "ZZZCLOSTRIDIUM SPECIES": 307, "ZZZENTEROBACTER AEROGENES": 308, "ZZZMYCOBACTERIUM MUCOGENICUM": 309, "ZZZPROPIONIBACTERIUM ACNES": 310, "ZZZPROTEUS PENNERI": 311, "ZZZPROTEUS VULGARIS": 312, "ZZZSTREPTOCOCCUS GALLOLYTICUS SUBSP. GALLOLYTICUS": 313, "ZZZSTREPTOCOCCUS GALLOLYTICUS SUBSP. PASTEURIANUS": 314}

label_encoders/susceptibility_label_encoder.json

@@ -0,0 +1 @@
+{"Inconclusive": 0, "Intermediate": 1, "Null": 2, "Resistant": 3, "Susceptible": 4, "Synergism": 5}

prediction.py

@@ -0,0 +1,65 @@
+# prediction.py
+
+import pandas as pd
+from utils import extract_genus, load_rules
+import re
+
+rules = load_rules()
+
+def encode_input(input_data, encoders):
+    """Encode input data using label encoders."""
+    encoded_data = {}
+    for col, value in input_data.items():
+        if col in encoders:
+            try:
+                encoded_data[col] = encoders[col][value]
+            except KeyError:
+                return f"Error: '{value}' is not a valid option for '{col}'"
+        else:
+            encoded_data[col] = value  # Numeric values pass through
+    return encoded_data
+
+def decode_output(encoded_value, encoder_name, encoders):
+    """Decode encoded values back to original labels."""
+    if encoder_name in encoders:
+        reverse_mapping = {v: k for k, v in encoders[encoder_name].items()}
+        return reverse_mapping.get(encoded_value, "Unknown")
+    return "Unknown"
+
+def predict_susceptibility(input_data, model, encoders):
+    """Predict susceptibility using the model and rules."""
+    try:
+        organism = input_data["organism"]
+        antibiotic = input_data["antibiotic"]
+        genus = extract_genus(organism)
+
+        # Check if any rule matches using regex
+        matching_rule = next((rules[key] for key in rules if re.match(fr"^{key[0]}", genus) and key[1] == antibiotic), None)
+
+        rule_output = matching_rule if matching_rule else "No specific rule found"
+
+        # Model prediction
+        encoded_data = encode_input(input_data, encoders)
+        if isinstance(encoded_data, str):  
+            return encoded_data
+
+        encoded_df = pd.DataFrame([encoded_data])
+        model_prediction = model.predict(encoded_df)[0]
+        model_output = decode_output(model_prediction, "susceptibility", encoders)
+
+        # Apply rule override if needed
+        if rule_output != "No specific rule found" and rule_output != model_output:
+            final_output = rule_output
+            reason = "Rule-based correction applied."
+        else:
+            final_output = model_output
+            reason = "Model prediction used."
+
+        return {
+            "Final Output": final_output,
+            "Rule Guidance": rule_output,
+            "Model Prediction": model_output,
+            "Decision Reason": reason
+        }
+    except Exception as e:
+        return {"Error": str(e)}

requirements.txt

@@ -0,0 +1,6 @@
+streamlit
+scikit-learn==1.2.2
+pandas
+numpy
+joblib
+regex

utils.py

@@ -0,0 +1,32 @@
+# utils.py
+
+import joblib
+import json
+import os
+import re
+
+def load_model(model_path):
+    """Load the trained model."""
+    return joblib.load(model_path)
+
+def load_label_encoders(encoder_dir):
+    """Load label encoders from JSON files."""
+    encoders = {}
+    for filename in os.listdir(encoder_dir):
+        if filename.endswith('_label_encoder.json'):
+            column = filename.replace('_label_encoder.json', '')
+            with open(os.path.join(encoder_dir, filename), 'r') as file:
+                encoders[column] = json.load(file)
+    return encoders
+
+def load_rules():
+    """Load rule-based susceptibility rules."""
+    return {
+        ("ACINETOBACTER", "Aztreonam"): "Resistant",
+        ("ACINETOBACTER", "Cefazolin"): "Resistant",
+        # Add all other rules here...
+    }
+
+def extract_genus(organism_name):
+    """Extract the genus from the organism name."""
+    return organism_name.split()[0].upper()