{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/yw/qz00x75d7kb98f7vm8dkhkvw0000gn/T/ipykernel_57352/998829134.py:6: LangChainDeprecationWarning: As of langchain-core 0.3.0, LangChain uses pydantic v2 internally. The langchain_core.pydantic_v1 module was a compatibility shim for pydantic v1, and should no longer be used. Please update the code to import from Pydantic directly.\n", "\n", "For example, replace imports like: `from langchain_core.pydantic_v1 import BaseModel`\n", "with: `from pydantic import BaseModel`\n", "or the v1 compatibility namespace if you are working in a code base that has not been fully upgraded to pydantic 2 yet. \tfrom pydantic.v1 import BaseModel\n", "\n", " from OpenAITools.CrinicalTrialTools import SimpleClinicalTrialAgent, GraderAgent, LLMTranslator, generate_ex_question_English\n", "/Users/satoc/miniforge3/envs/gradio/lib/python3.12/site-packages/transformers/tokenization_utils_base.py:1617: FutureWarning: `clean_up_tokenization_spaces` was not set. It will be set to `True` by default. This behavior will be deprecated in transformers v4.45, and will be then set to `False` by default. For more details check this issue: https://github.com/huggingface/transformers/issues/31884\n", " warnings.warn(\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "* Running on local URL: http://127.0.0.1:7861\n", "\n", "To create a public link, set `share=True` in `launch()`.\n" ] }, { "data": { "text/html": [ "
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" }, { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/yw/qz00x75d7kb98f7vm8dkhkvw0000gn/T/ipykernel_57352/998829134.py:29: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " df['AgentJudgment'] = None\n", "/var/folders/yw/qz00x75d7kb98f7vm8dkhkvw0000gn/T/ipykernel_57352/998829134.py:30: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " df['AgentGrade'] = None\n" ] } ], "source": [ "import gradio as gr\n", "import pandas as pd\n", "from OpenAITools.FetchTools import fetch_clinical_trials\n", "from langchain_openai import ChatOpenAI\n", "from langchain_groq import ChatGroq\n", "from OpenAITools.CrinicalTrialTools import SimpleClinicalTrialAgent, GraderAgent, LLMTranslator, generate_ex_question_English\n", "from OpenAITools.JRCTTools import get_matched_df,GetJRCTCriteria\n", "from sentence_transformers import SentenceTransformer\n", "from sentence_transformers import util\n", "\n", "# モデルとエージェントの初期化\n", "groq = ChatGroq(model_name=\"llama3-70b-8192\", temperature=0)\n", "translator = LLMTranslator(groq)\n", "CriteriaCheckAgent = SimpleClinicalTrialAgent(groq)\n", "grader_agent = GraderAgent(groq)\n", "selectionModel = SentenceTransformer('pritamdeka/S-PubMedBert-MS-MARCO')\n", "\n", "# データフレームを生成する関数\n", "def generate_dataframe(age, sex, tumor_type, GeneMutation, Meseable, Biopsiable):\n", " # 日本語の腫瘍タイプを英語に翻訳\n", " TumorName = translator.translate(tumor_type)\n", "\n", " # 質問文を生成\n", " ex_question = generate_ex_question_English(age, sex, TumorName, GeneMutation, Meseable, Biopsiable)\n", " \n", " # 臨床試験データの取得\n", " basedf = pd.read_csv(\"../ClinicalTrialCSV/JRCT20241215CancerPost.csv\", index_col=0)\n", " df = get_matched_df(basedf=basedf, query=TumorName, model=selectionModel, threshold=0.925)\n", " df['AgentJudgment'] = None\n", " df['AgentGrade'] = None\n", " \n", " # 臨床試験の適格性の評価\n", " progress = gr.Progress(track_tqdm=True)\n", " for i in range(len(df)):\n", " TargetCriteria = GetJRCTCriteria(df, i)\n", " AgentJudgment = CriteriaCheckAgent.evaluate_eligibility(TargetCriteria, ex_question)\n", " AgentGrade = grader_agent.evaluate_eligibility(AgentJudgment)\n", " # df.locを使って値を代入（行・列名で指定）\n", " df.loc[df.index[i], 'AgentJudgment'] = AgentJudgment\n", " df.loc[df.index[i], 'AgentGrade'] = AgentGrade\n", " progress((i + 1) / len(df))\n", " \n", " # 列を指定した順に並び替え\n", " columns_order = ['JRCT ID', 'Title', '研究・治験の目的','AgentJudgment', 'AgentGrade','主たる選択基準', '主たる除外基準','Inclusion Criteria','Exclusion Criteria','NCT No', 'JapicCTI No']\n", " df = df[columns_order]\n", " \n", " return df, df # フィルタ用と表示用にデータフレームを返す\n", "\n", "# 特定のAgentGrade（yes, no, unclear）に基づいて行をフィルタリングする関数\n", "def filter_rows_by_grade(original_df, grade):\n", " df_filtered = original_df[original_df['AgentGrade'] == grade]\n", " return df_filtered, df_filtered\n", "\n", "# CSVとして保存しダウンロードする関数\n", "def download_filtered_csv(df):\n", " file_path = \"filtered_data.csv\"\n", " df.to_csv(file_path, index=False)\n", " return file_path\n", "\n", "# 全体結果をCSVとして保存しダウンロードする関数\n", "def download_full_csv(df):\n", " file_path = \"full_data.csv\"\n", " df.to_csv(file_path, index=False)\n", " return file_path\n", "\n", "# Gradioインターフェースの作成\n", "with gr.Blocks() as demo:\n", " gr.Markdown(\"## 臨床試験適格性評価インターフェース\")\n", "\n", " # 各種入力フィールド\n", " age_input = gr.Textbox(label=\"Age\", placeholder=\"例: 65\")\n", " sex_input = gr.Dropdown(choices=[\"男性\", \"女性\"], label=\"Sex\")\n", " tumor_type_input = gr.Textbox(label=\"Tumor Type\", placeholder=\"例: gastric cancer, 日本でも良いですが英語の方が精度が高いです。\")\n", " gene_mutation_input = gr.Textbox(label=\"Gene Mutation\", placeholder=\"例: HER2\")\n", " measurable_input = gr.Dropdown(choices=[\"有り\", \"無し\", \"不明\"], label=\"Measurable Tumor\")\n", " biopsiable_input = gr.Dropdown(choices=[\"有り\", \"無し\", \"不明\"], label=\"Biopsiable Tumor\")\n", "\n", " # データフレーム表示エリア\n", " dataframe_output = gr.DataFrame()\n", " original_df = gr.State()\n", " filtered_df = gr.State()\n", "\n", " # データフレーム生成ボタン\n", " generate_button = gr.Button(\"Generate Clinical Trials Data\")\n", "\n", " # フィルタリングボタン\n", " yes_button = gr.Button(\"Show Eligible Trials\")\n", " no_button = gr.Button(\"Show Ineligible Trials\")\n", " unclear_button = gr.Button(\"Show Unclear Trials\")\n", " \n", " # ダウンロードボタン\n", " download_filtered_button = gr.Button(\"Download Filtered Data\")\n", " download_filtered_output = gr.File(label=\"Download Filtered Data\")\n", "\n", " download_full_button = gr.Button(\"Download Full Data\")\n", " download_full_output = gr.File(label=\"Download Full Data\")\n", "\n", "\n", " # ボタン動作の設定\n", " generate_button.click(fn=generate_dataframe, inputs=[age_input, sex_input, tumor_type_input, gene_mutation_input, measurable_input, biopsiable_input], outputs=[dataframe_output, original_df])\n", " yes_button.click(fn=filter_rows_by_grade, inputs=[original_df, gr.State(\"yes\")], outputs=[dataframe_output, filtered_df])\n", " no_button.click(fn=filter_rows_by_grade, inputs=[original_df, gr.State(\"no\")], outputs=[dataframe_output, filtered_df])\n", " unclear_button.click(fn=filter_rows_by_grade, inputs=[original_df, gr.State(\"unclear\")], outputs=[dataframe_output, filtered_df])\n", " download_filtered_button.click(fn=download_filtered_csv, inputs=filtered_df, outputs=download_filtered_output)\n", " download_full_button.click(fn=download_full_csv, inputs=original_df, outputs=download_full_output)\n", "\n", "\n", "# インターフェースの起動\n", "demo.launch()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "gradio", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.3" } }, "nbformat": 4, "nbformat_minor": 2 }