app.py

import gradio as gr
import openai
from openai import OpenAI
import os
import io
import base64

# Set API key and organization ID from environment variables
api_key = os.environ.get("OPENAI_API_KEY")
#base_url = os.environ.get("OPENAI_API_BASE")
client = OpenAI(api_key=api_key)

# Define the model to be used
MODEL = os.environ.get("MODEL")

def read(filename):
    with open(filename) as f:
        data = f.read()
    return data
    
SYS_PROMPT = read('system_prompt.txt')


DESCRIPTION = '''
<div>
<h1 style="text-align: center;">诊疗助手Alpha</h1>
<p>一个帮助您分析症状和检验报告的AI工具。</p>
<p>🔎 选择您需要咨询的科室，在输入框中输入症状描述或者体检信息等；您也可以在图片框中上传检测报告图。</p>
<p>🦕 请注意生成信息可能不准确，且不具备任何实际参考价值，如有需要请联系专业医生。</p>
</div>
'''


css = """
h1 {
    text-align: center;
    display: block;
}
footer {
    display:none !important
}
"""


LICENSE = '采用 ' + MODEL + ' 模型'
    
def process_text(text_input, unit):
    if text_input:
        completion = client.chat.completions.create(
            model=MODEL,
            messages=[
                {"role": "system", "content": f" You are a experienced {unit} doctor." + SYS_PROMPT},
                {"role": "user", "content": f"Hello! Could you solve {text_input}?"}
            ]
        )
        return completion.choices[0].message.content
    return ""

def encode_image_to_base64(image_input):
    buffered = io.BytesIO()
    image_input.save(buffered, format="JPEG")
    img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
    return img_str

def process_image(image_input, unit):
    if image_input is not None:
        #with open(image_input.name, "rb") as f:
        #    base64_image = base64.b64encode(f.read()).decode("utf-8")
        base64_image = encode_image_to_base64(image_input)
        response = client.chat.completions.create(
            model=MODEL,
            messages=[
                {"role": "system", "content": f" You are a experienced {unit} doctor." + SYS_PROMPT},
                {"role": "user", "content": [
                    {"type": "text", "text": "Help me understand what is in this picture and analysis."},
                    {"type": "image_url", 
                     "image_url": {
                        "url": f"data:image/jpeg;base64,{base64_image}",
                        "detail":"low"}
                    }
                ]}
            ],
            temperature=0.0,
            max_tokens=1024,
        )
        return response.choices[0].message.content


def main(text_input="", image_input=None, unit=""):
    if text_input and image_input is None:
        return process_text(text_input,unit)
    elif image_input is not None:
        return process_image(image_input,unit)

with gr.Blocks(theme='shivi/calm_seafoam', css=css, title="诊疗助手Alpha") as iface:
    with gr.Accordion(""):
        gr.Markdown(DESCRIPTION)
        unit = gr.Dropdown(label="🩺科室", value='中医科', elem_id="units",
                            choices=["中医科", "内科", "外科", "妇产科",  "儿科", \
                                     "五官科", "男科", "皮肤性病科", "传染科", "精神心理科", \
                                        "整形美容科", "营养科", "生殖中心", "麻醉医学科", "医学影像科", \
                                            "骨科", "肿瘤科", "急诊科", "检验科"])
    with gr.Row():
        output_box = gr.Markdown(label="分析")  # Create an output textbox
    with gr.Row():
        image_input = gr.Image(type="pil", label="上传图片")  # Create an image upload button
        text_input = gr.Textbox(label="输入")  # Create a text input box
    with gr.Row():
        submit_btn = gr.Button("🚀 确认")  # Create a submit button
        clear_btn = gr.ClearButton(output_box, value="🗑️ 清空") # Create a clear button

    # Set up the event listeners
    submit_btn.click(main, inputs=[text_input, image_input, unit], outputs=output_box)

    gr.Markdown(LICENSE)
    
#gr.close_all()

iface.queue().launch(show_api=False)  # Launch the Gradio interface