app.py

import gradio as gr
import os
from typing import Tuple, Optional
import os
import shutil
import sys
from pathlib import Path
import cv2
import gradio as gr
import numpy as np
import spaces
# import supervision as sv
import torch
from PIL import Image
from tqdm import tqdm
import sys
from pathlib import Path
from huggingface_hub import login
# from dotenv import load_dotenv
from transformers import AutoModel, AutoTokenizer, BitsAndBytesConfig

# For Hugging Face Spaces, secrets are automatically loaded as environment variables
token = os.getenv("HF_TOKEN")
if token:
    login(token=token)
# Clear Hugging Face cache
# cache_dirs = [
#     "/home/user/.cache/huggingface/",
#     "/home/user/.cache/torch/",
#     "/home/user/.cache/pip/"
# ]

# for cache_dir in cache_dirs:
#     if os.path.exists(cache_dir):
#         print(f"Clearing cache: {cache_dir}")
#         shutil.rmtree(cache_dir, ignore_errors=True)
# Add the current directory to Python path
current_dir = Path(__file__).parent
sys.path.append(str(current_dir))
# sys.path.append("./BiomedParse/")
# BIOMEDPARSE_PATH = Path(__file__).parent / "BiomedParse"
# sys.path.append(str(BIOMEDPARSE_PATH))
# sys.path.append(str(BIOMEDPARSE_PATH / "BiomedParse"))  # Add the inner BiomedParse directory
from modeling.BaseModel import BaseModel
from modeling import build_model
from utilities.arguments import load_opt_from_config_files
from utilities.constants import BIOMED_CLASSES
from inference_utils.inference import interactive_infer_image
from inference_utils.output_processing import check_mask_stats
from inference_utils.processing_utils import read_rgb

import spaces

# breakpoint()
MARKDOWN = """
<div align="center" style="padding: 20px 0;">
    <h1 style="font-size: 3em; margin: 0;">
        ሀ<span style="color: #32CD32;">A</span>ኪ<span style="color: #FFD700;">i</span>ም
        <sup style="font-size: 0.5em;">AI</sup>
    </h1>

    <div style="display: flex; justify-content: center; align-items: center; gap: 15px; margin: 15px 0;">
        <a href="https://cyberbrainai.com/">
            <img src="https://cyberbrainai.com/assets/logo.svg" alt="CyberBrain AI" style="width:40px; height:40px; vertical-align: middle;">
        </a>
        <a href="https://colab.research.google.com/drive/1p3Yf_6xdZPMz5RUtt_NyxrDjrbSgvTDy#scrollTo=t30NqIrCKdAI">
            <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="ድinቅneሽ" style="vertical-align: middle;">
        </a>
        <a href="https://www.youtube.com/watch?v=Dv003fTyO-Y">
            <img src="https://badges.aleen42.com/src/youtube.svg" alt="YouTube" style="vertical-align: middle;">
        </a>
    </div>
</div>
    <div>
        <p style="font-size: 1.4em; line-height: 1.5; margin: 15px 0; text-align: left;">
            This demo integrates BiomedParse, a foundation model for joint segmentation, detection, and recognition across 9 biomedical imaging modalities. 
            The model supports <span style="color: #FF4500;">CT</span>, <span style="color: #4169E1;">MRI</span>, <span style="color: #32CD32;">X-Ray</span>, <span style="color: #9370DB;">Pathology</span>, <span style="color: #FFD700;">Ultrasound</span>, <span style="color: #FF69B4;">Endoscope</span>, <span style="color: #20B2AA;">Fundus</span>, <span style="color: #FF8C00;">Dermoscopy</span>, and <span style="color: #8B008B;">OCT</span>.
        </p>
    </div>

"""

IMAGE_PROCESSING_EXAMPLES = [
    ["BiomedParse Segmentation", 
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/T0011.jpg",
     "Optic disc in retinal Fundus"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/Part_3_226_pathology_breast.png",
     "optic disc, optic cup"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/covid_1585.png",
     "COVID-19 infection in chest X-Ray"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/TCGA_HT_7856_19950831_8_MRI-FLAIR_brain.png",
     "Lower-grade glioma in brain MRI"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/LIDC-IDRI-0140_143_280_CT_lung.png",
     "COVID-19 infection in chest CT"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/144DME_as_F.jpeg",
     "Cystoid macular edema in retinal OCT"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/Part_1_516_pathology_breast.png",
     "Glandular structure in colon Pathology"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/ISIC_0015551.jpg",
     "Melanoma in skin Dermoscopy"],
    ["BiomedParse Segmentation",
     "https://raw.githubusercontent.com/microsoft/BiomedParse/main/examples/C3_EndoCV2021_00462.jpg",
     "Neoplastic polyp in colon Endoscope"]
]

BIOMEDPARSE_MODES = {
    "CT-Abdomen": ["abdomen", "liver"],
    "CT-Chest": ["lung"],
    "CT-Liver": ["liver"],
    "MRI-Abdomen": ["abdomen"],
    "MRI-Cardiac": ["heart"],
    "MRI-FLAIR-Brain": ["brain"],
    "MRI-T1-Gd-Brain": ["brain"],
    "Pathology": ["bladder", "breast", "cervix", "colon", "esophagus", "kidney", 
                  "liver", "ovarian", "prostate", "stomach", "testis", "thyroid", "uterus"],
    "X-Ray-Chest": ["chest"],
    "Ultrasound-Cardiac": ["heart"],
    "Endoscopy": ["colon"],
    "Fundus": ["retinal"],
    "Dermoscopy": ["skin"],
    "OCT": ["retinal"]
}

IMAGE_INFERENCE_MODES = [
    "BIOMED SEGMENTATION",
    "BIOMED DETECTION", 
    "BIOMED RECOGNITION",
    "BIOMED SEGMENTATION + DETECTION",
    "BIOMED SEGMENTATION + RECOGNITION",
    "BIOMED DETECTION + RECOGNITION",
    "BIOMED SEGMENTATION + DETECTION + RECOGNITION"
]

MODALITY_PROMPTS = {
   "CT-Abdomen": ["postcava", "aorta", "right kidney", "kidney", "left kidney", "duodenum", "pancreas", "liver", "spleen", "stomach", "gallbladder", "left adrenal gland", "adrenal gland", "right adrenal gland", "esophagus"],
   "CT-Chest": ["nodule", "COVID-19 infection", "tumor"],
   "MRI-Abdomen": ["aorta", "postcava", "right kidney", "duodenum", "kidney", "left kidney", "liver", "pancreas", "gallbladder", "stomach", "spleen", "left adrenal gland", "adrenal gland", "right adrenal gland", "esophagus"],
   "MRI-Cardiac": ["left heart ventricle", "myocardium", "right heart ventricle"],
   "MRI-FLAIR-Brain": ["edema", "tumor core", "whole tumor"],
   "MRI-T1-Gd-Brain": ["enhancing tumor", "non-enhancing tumor", "tumor core"],
   "Pathology": ["connective tissue cells", "inflammatory cells", "neoplastic cells", "epithelial cells"],
   "X-Ray-Chest": ["left lung", "lung", "right lung"],
   "Ultrasound-Cardiac": ["left heart atrium", "left heart ventricle"],
   "Endoscopy": ["neoplastic polyp", "polyp", "non-neoplastic polyp"],
   "Fundus": ["optic cup", "optic disc"],
   "Dermoscopy": ["lesion", "melanoma"],
   "OCT": ["edema"] }


def on_mode_dropdown_change(selected_mode):
    if selected_mode in IMAGE_INFERENCE_MODES:
        # Show modality dropdown and hide other inputs initially
        return [
            gr.Dropdown(visible=True, choices=list(BIOMEDPARSE_MODES.keys()), label="Modality"),
            gr.Dropdown(visible=True, label="Anatomical Site"),
            gr.Textbox(visible=False),
            gr.Textbox(visible=False)
        ]
    else:
        # Original behavior for other modes
        return [
            gr.Dropdown(visible=False),
            gr.Dropdown(visible=False),
            gr.Textbox(visible=True),
            gr.Textbox(visible=(selected_mode == None))
        ]

def on_modality_change(modality):
    if modality:
        return gr.Dropdown(choices=BIOMEDPARSE_MODES[modality], visible=True)
    return gr.Dropdown(visible=False)


def initialize_model():
    opt = load_opt_from_config_files(["configs/biomedparse_inference.yaml"])
    pretrained_pth = 'hf_hub:microsoft/BiomedParse'
    opt['device'] = 'cuda' if torch.cuda.is_available() else 'cpu'
    model = BaseModel(opt, build_model(opt)).from_pretrained(pretrained_pth).eval()
    with torch.no_grad():
        model.model.sem_seg_head.predictor.lang_encoder.get_text_embeddings(
            BIOMED_CLASSES + ["background"], is_eval=True
        )
    return model

def initialize_llm():
    try:
        print("Starting LLM initialization...")
        # Add quantization config
        quantization_config = BitsAndBytesConfig(
            load_in_4bit=True,
            bnb_4bit_compute_dtype=torch.float16
        )
        
        model = AutoModel.from_pretrained(
            "ContactDoctor/Bio-Medical-MultiModal-Llama-3-8B-V1",
            device_map="auto",
            torch_dtype=torch.float16,
            trust_remote_code=True,
            low_cpu_mem_usage=True,
            quantization_config=quantization_config
        )
        print("Model loaded successfully")
        
        tokenizer = AutoTokenizer.from_pretrained(
            "ContactDoctor/Bio-Medical-MultiModal-Llama-3-8B-V1",
            trust_remote_code=True
        )
        print("Tokenizer loaded successfully")
        return model, tokenizer
    except Exception as e:
        print(f"Failed to initialize LLM: {str(e)}")
        return None, None

model = initialize_model()
llm_model, llm_tokenizer = initialize_llm()

def update_example_prompts(modality):
    if modality in MODALITY_PROMPTS:
        examples = MODALITY_PROMPTS[modality]
        return f"Example prompts for {modality}:\n" + ", ".join(examples)
    return ""

# Utility functions
@spaces.GPU
@torch.inference_mode()
@torch.autocast(device_type="cuda", dtype=torch.bfloat16)
def process_image(image_path, text_prompts, modality):
    try:
        # Input validation
        if not image_path:
            raise ValueError("Please upload an image")
        if not text_prompts or text_prompts.strip() == "":
            raise ValueError("Please enter prompts for analysis")
        if not modality:
            raise ValueError("Please select a modality")

        # Original BiomedParse processing
        image = read_rgb(image_path)
        text_prompts = [prompt.strip() for prompt in text_prompts.split(',')]
        pred_masks = interactive_infer_image(model, Image.fromarray(image), text_prompts)
        
        # Prepare outputs
        results = []
        analysis_results = []
        
        # Process with BiomedParse
        for i, prompt in enumerate(text_prompts):
            p_value = check_mask_stats(image, pred_masks[i] * 255, modality, prompt)
            analysis_results.append(f"P-value for '{prompt}' ({modality}): {p_value:.4f}")
            
            overlay_image = image.copy()
            overlay_image[pred_masks[i] > 0.5] = [255, 0, 0]
            results.append(overlay_image)
        
        # Process with LLM only if available
        if llm_model is not None and llm_tokenizer is not None:
            print("LLM model and tokenizer are available")
            try:
                pil_image = Image.fromarray(image)
                question = 'Give the modality, organ, analysis, abnormalities (if any), treatment (if abnormalities are present)?'
                msgs = [{'role': 'user', 'content': [pil_image, question]}]
                
                print("Starting LLM inference...")
                llm_response = ""
                for new_text in llm_model.chat(
                    image=pil_image,
                    msgs=msgs,
                    tokenizer=llm_tokenizer,
                    sampling=True,
                    temperature=0.95,
                    stream=True
                ):
                    llm_response += new_text
                print(f"LLM generated response: {llm_response}")
                
                # Make the combined analysis more visible
                combined_analysis = "\n\n" + "="*50 + "\n"
                combined_analysis += "BiomedParse Analysis:\n"
                combined_analysis += "\n".join(analysis_results)
                combined_analysis += "\n\n" + "="*50 + "\n"
                combined_analysis += "LLM Analysis:\n"
                combined_analysis += llm_response
                combined_analysis += "\n" + "="*50
                
            except Exception as e:
                print(f"LLM analysis failed with error: {str(e)}")
                combined_analysis = "\n".join(analysis_results)
        else:
            print("LLM model or tokenizer is not available")
            combined_analysis = "\n".join(analysis_results)
        
        return results, combined_analysis

    except Exception as e:
        error_msg = f"⚠️ An error occurred: {str(e)}"
        print(f"Error details: {str(e)}", flush=True)
        return None, error_msg

# Define Gradio interface
with gr.Blocks() as demo:
    gr.HTML(MARKDOWN)    
    with gr.Row():        
        with gr.Column():
            image_input = gr.Image(type="filepath", label="Input Image")
            prompts_input = gr.Textbox(
                lines=2, 
                placeholder="Enter prompts separated by commas...", 
                label="Prompts"
            )
            modality_dropdown = gr.Dropdown(
                choices=list(BIOMEDPARSE_MODES.keys()),
                value=list(BIOMEDPARSE_MODES.keys())[0],
                label="Modality"
            )
            submit_btn = gr.Button("Submit")
        with gr.Column():
            output_gallery = gr.Gallery(label="Findings")
            pvalue_output = gr.Textbox(
                label="Results", 
                interactive=False,
                show_label=True
            )
            with gr.Accordion("Example Prompts by Modality", open=False):
                for modality, prompts in MODALITY_PROMPTS.items():
                    prompt_str = ", ".join(prompts)
                    gr.Markdown(f"**{modality}**: {prompt_str}")
    # Add error handling for the submit button
    submit_btn.click(
        fn=process_image,
        inputs=[image_input, prompts_input, modality_dropdown],
        outputs=[output_gallery, pvalue_output],
        api_name="process"
    )

demo.launch()