{ "cells": [ { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import torch\n", "from transformers import AutoTokenizer, AutoModel\n", "import re" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# Target列を分割する関数\n", "def split_target(target):\n", " # 指定された区切り文字で分割\n", " split_words = re.split(r'[,\\n、・及びおよび又はまたは]+', target)\n", " # 空白文字を除外してリストとして返す\n", " return [word.strip() for word in split_words if word.strip()]" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "basedf = pd.read_csv('../ClinicalTrialCSV/JRCT20241202Cancer.csv', index_col=0)\n", "basedf = basedf.dropna(subset=['試験等のフェーズ'])\n", "# Target列を分割してTargetWord列を追加\n", "basedf['TargetWord'] = basedf['Target'].apply(split_target)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "9671f824f52348f8a161a130e7437adb", "version_major": 2, "version_minor": 0 }, "text/plain": [ "tokenizer_config.json: 0%| | 0.00/418 [00:00\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
JRCT IDNCT NoJapicCTI NoTitleTarget研究・治験の目的試験等のフェーズ試験の種類無作為化盲検化...purposeInclusion CriteriaExclusion CriteriaAge MinimumAge MaximumGenderDiscontinuation CriteriaKeywordIntervention(s)TargetWord
12jRCT2041240130NaNNaN中等度-高度PONV発症リスクを有する肺切除術予定患者の周術期管理におけるGranisetr...肺がん全身麻酔を要する肺切除術周術期における標準制吐療法であるグラニセトロン(Granisetro...2NaN無作為化比較非盲検...treatment purpose(1) Patients scheduled to undergo anatomical l...(1) Patients with a history of allergy to the ...18age old overNo limitBothNaNNaNTreatment group\\r\\nAdminister GRA 1mg intraven...[肺がん]
59jRCT2031240404NaNNaN転移性HR+/HER2-乳がん及びその他の進行性固形がん患者を対象にCDK4阻害剤BGB-4...・進行性固形腫瘍・進行性乳がん・転移性乳がん・ホルモン受容体陽性乳がん・ホルモン受容体陽性乳...サイクリン依存性キナーゼ 4 (CDK4) 阻害剤である BGB-43395 が、ホルモン受...1NaN非無作為化比較非盲検...treatment purpose-Phase 1a (Dose Escalation) and 1b (Dose Expan...-Prior therapy selectively targeting CDK4 (pri...18age old overNo limitBothNaNNaN-Drug: BGB-43395\\r\\nPlanned doses administered...[進行性固形腫瘍, 進行性乳がん, 転移性乳がん, ホルモン受容体陽性乳がん, ホルモン受容...
201jRCT2052240059NaNNaN遺伝子HSD17B4高メチル化を有するHER2陽性ER陰性乳癌における非手術療法の有用性を評...乳がんHSD17B4高メチル化（HSD17B4 hypermethylation：HH）を有するH...2NaN単一群非盲検...diagnostic purpose1. Histologically confirmed invasive breast ca...1. History of other malignancy within the last...20age old overNo limitFemaleNaNNaNOmitting breast surgery after preoperative che...[乳がん]
234jRCT2031240065NaNNaNDNAミスマッチ修復機構欠損を有する進行性・転移性子宮体がん患者を対象に、一次治療として化学...子宮体がんdMMR/MSI-Hを有する再発又は進行子宮体癌患者を対象にdostarlimabの有効性及...3NaN無作為化比較非盲検...treatment purpose1. Female patient is at least 18 years of age,...1. Patient has received neoadjuvant/adjuvant s...18age old overNo limitFemaleNaNNaNPatients will be randomized 1:1 to receive eit...[子宮体がん]
257jRCT2033240023NaNNaN免疫チェックポイント阻害剤に対してacquired resistanceとなったがん患者を対...肺がん、食道がん、胃癌等の固形がん免疫チェックポイント阻害剤に対してacquired resistanceとなったがん患者（非...1-2NaN単一群非盲検...treatment purpose1. Japanese patients aged 20 years or older at...1. Patients with a history of severe drug hype...20age old overNo limitBothNaNNaNIntramuscular injection of the investigational...[肺がん, 食道がん, 胃癌等の固形がん]
271jRCT2031230750NCT06188559NaN治療歴のある HER2 陽性又は HER2 低発現の切除不能又は転移性乳癌の被験者を対象に，...HER2 陽性又は HER2 低発現の切除不能又は転移性乳癌治療2NaN無作為化比較非盲検...treatment purpose(1) Male or female, aged >=18 years at the tim...(1) Presence of brain or subdural metastases, ...18age old overNo limitBothNaNNaNGeneric Name:NA\\r\\nStudy Treatment in dose opt...[HER2 陽性, HER2 低発現の切除不能, 転移性乳癌]
279jRCT2031230723NCT06112379NaN未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、...乳癌未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、...3NaN無作為化比較非盲検...treatment purpose1. Participant must be >= 18 years at the time...1. As judged by the investigator, any evidence...18age old overNo limitBothNaNNaN- Experimental arm: Dato-DXd plus durvalumab n...[乳癌]
311jRCT2061230102NCT06103864NaNProgrammed death-ligand（PD-L1）陽性の局所再発手術不能又は転移性...乳癌PD-L1陽性の局所再発手術不能または転移性TNBC患者を対象に、デュルバルマブ併用または非...3NaN無作為化比較非盲検...treatment purposeHistologically or cytologically documented loc...As judged by investigator, severe or uncontrol...18age old overNo limitBothNaNNaNArm 1: Dato-DXd + durvalumab\\r\\nArm 2: Investi...[乳癌]
396jRCT2061230074NCT05952557NaN根治的局所治療（化学療法の併用または非併用）を受けて疾患の兆候のない、再発リスクが中間～高リ...乳がん、早期乳がんTreatment3NaN無作為化比較非盲検...treatment purpose- Women and Men; 18 years or more at the time ...- Inoperable locally advanced or metastatic br...18age old overNo limitNaNNaNNaNarm A: continue with SoC ET as directed by inv...[乳がん, 早期乳がん]
467jRCT2031230249NCT05753501NaN[M23-647]B細胞性悪性腫瘍患者を対象としたBTK 分解誘導化合物ABBV-101 の...血液がん・ABBV-101 の安全性を明らかにし，忍容性を確認する\\r\\n・ABBV-101 の経口...1NaN単一群非盲検...treatment purpose-For Dose Escalation (Part 1) only: Participan...-Previously treated with a Bruton's tyrosine k...18age old overNo limitBothNaNNaNDrug: ABBV-101\\r\\nOral: Tablet[血液がん]
509jRCT2031230109NCT05514054NaNEMBER-4：2～5年間の術後内分泌療法による前治療歴を有する再発高リスクのER+、HER...乳癌早期乳癌患者を対象としたimlunestrantと標準的な内分泌療法の比較試験3NaN無作為化比較非盲検...treatment purpose-Have a diagnosis of ER+, HER2- early-stage, r...-Have any evidence of metastatic disease (incl...18age old overNo limitBothNaNNaN-Drug: Imlunestrant\\r\\n Administered orally.\\...[乳癌]
514jRCT2031230090NaNNaNMELK阻害剤OTS167POにおける転移性・進行性乳がん患者を対象とした安全性、忍容性およ...再発・難治性の局所進行性・転移性乳がん及びトリプルネガティブ乳がん再発・難治性の局所進行性または転移性乳がん患者に対して、OTS167を経口カプセルで投与する...1NaN単一群非盲検...treatment purposeDose Escalation and Dose Expansion Cohorts\\r\\n...Dose Escalation and Expansion Cohorts\\r\\n1.\\tW...18age old overNo limitFemaleNaNNaNThis is a Phase I dose escalation/expansion mu...[再発, 難治性の局所進行性, 転移性乳がん, トリプルネガティブ乳がん]
515jRCT2031230096NCT05774951NaN根治的局所領域療法（化学療法の併用または非併用）および標準補助内分泌療法（ET）を少なくとも...乳がん、早期乳がんTreatment3NaN無作為化比較非盲検...treatment purpose- Women and Men, greater than or equal to 18 y...- Inoperable locally advanced or metastatic br...18age old overNo limitBothNaNNaNarm A: continue with SoC ET as directed by inv...[乳がん, 早期乳がん]
582jRCT2051220176NaNNaN切除不能進行再発食道がん及び胃がんに対する免疫チェックポイント阻害剤(ICI)と光線力学的療...食道がん、胃がん一次治療として免疫チェックポイント阻害剤（ICI）を含む標準治療を予定される進行・再発食道が...2NaN単一群非盲検...treatment purposeThe patient with unresectable or recurrent eso...Has been diagnosed with HER2-positive gastric ...18age old overNo limitBothNaNEsophageal cancer, Gastric cancerCohort A: Nivolumab(240mg, every 2 weeks or 36...[食道がん, 胃がん]
610jRCT2061220087NCT05629585NaN術前薬物療法後の外科的切除時に乳房及び／又は腋窩リンパ節に浸潤性残存病変を有するステージI～...乳癌術前薬物療法後の外科的切除時に乳房および／または腋窩リンパ節に浸潤性残存病変を有するI～II...3NaN無作為化比較非盲検...treatment purposeParticipant must be >= 18 years at the time of...Stage IV (metastatic) TNBC.\\r\\nHistory of prio...18age old over130age old underBothNaNNaNArm 1: Dato-DXd 6 mg/kg IV Q3W x 8 cycles + Du...[乳癌]
693jRCT2031220276NCT05307705NaNPIK3CA H1047R変異を有する進行乳がん患者及びその他の固形がん患者を対象としたLO...乳がんLOXO-783 の単独投与及び他の抗がん剤との併用投与における 第 2 相試験の推奨用量、...1NaN単一群非盲検...treatment purpose-Have advanced breast cancer or another solid ...-Medical Conditions\\r\\n -Colorectal cancer\\r\\n...18age old overNo limitBothNaNNaN-Drug: LOXO-783\\r\\n Oral\\r\\n Other Name: LY384...[乳がん]
708jRCT2031220203NaNNaN初回腫瘍減量手術肉眼的完全切除後の進行卵巣癌を対象にパクリタキセル、カルボプラチン投与後のニ...卵巣がん診断時にステージIIIA/B/Cの高悪性度上皮性卵巣癌で、PDS後に残存病変を認めない患者を...2NaN無作為化比較非盲検...treatment purpose1. Female patient >= 18 years of age.\\r\\n2. Si...1. Patient with clear cell adenocarcinoma or c...18age old overNo limitFemaleNaNNaNPatients will be included after surgery and wi...[卵巣がん]
753jRCT2031220034NCT04862663NaN局所進行（切除不能）又は転移性のホルモン受容体陽性／ヒト上皮細胞成長因子受容体2 陰性乳癌を...局所進行（切除不能）又は転移性乳癌Treatment3NaN無作為化比較二重盲検...treatment purposeKey inclusion criteria for both phases:\\n\\r\\n1...Key exclusion criteria for both phases:\\r\\n1. ...18age old over99age old underBothNaNNaNCapivasertib: PO BD 4 days on /3 days off per ...[局所進行（切除不能）, 転移性乳癌]
802jRCT2031210585NCT05123482NaN進行又は転移性固形がん患者を対象としたAZD8205の単独療法及び他の抗がん剤との併用療法に...乳癌、胆道癌、卵巣癌、子宮内膜癌進行又は転移性固形がんに対する治療法候補としての新規化合物AZD8205の研究1-2NaN単一群非盲検...treatment purpose- Age18 years or more\\n\\r\\n- Relapsed/metastat...- Treatment with any of the following:\\n\\r\\n1....18age old overNo limitBothNaNNaNSub-study1\\r\\nAZD8205 Monotherapy\\r\\nAZD8205 i...[乳癌, 胆道癌, 卵巣癌, 子宮内膜癌]
865jRCT2052210099NaNNaN乳がんを有する成人女性及び健康成人女性を対象とした乳房用マイクロ波画像診断装置IGS-000...乳がん本治験の目的は，乳がん又はその疑いのある者，並びに乳がん又はその疑いがない者を対象に乳房画像...2NaN単一群単盲検...screeningBreast cancer (including suspected) \\r\\n(1) On...(1) There is trauma with bleeding on the breas...20age old overNo limitFemaleNaNNaNThe test using IGS-0001 will be performed twic...[乳がん]
\n", "

20 rows × 39 columns

\n", "" ], "text/plain": [ " JRCT ID NCT No JapicCTI No \\\n", "12 jRCT2041240130 NaN NaN \n", "59 jRCT2031240404 NaN NaN \n", "201 jRCT2052240059 NaN NaN \n", "234 jRCT2031240065 NaN NaN \n", "257 jRCT2033240023 NaN NaN \n", "271 jRCT2031230750 NCT06188559 NaN \n", "279 jRCT2031230723 NCT06112379 NaN \n", "311 jRCT2061230102 NCT06103864 NaN \n", "396 jRCT2061230074 NCT05952557 NaN \n", "467 jRCT2031230249 NCT05753501 NaN \n", "509 jRCT2031230109 NCT05514054 NaN \n", "514 jRCT2031230090 NaN NaN \n", "515 jRCT2031230096 NCT05774951 NaN \n", "582 jRCT2051220176 NaN NaN \n", "610 jRCT2061220087 NCT05629585 NaN \n", "693 jRCT2031220276 NCT05307705 NaN \n", "708 jRCT2031220203 NaN NaN \n", "753 jRCT2031220034 NCT04862663 NaN \n", "802 jRCT2031210585 NCT05123482 NaN \n", "865 jRCT2052210099 NaN NaN \n", "\n", " Title \\\n", "12 中等度-高度PONV発症リスクを有する肺切除術予定患者の周術期管理におけるGranisetr... \n", "59 転移性HR+/HER2-乳がん及びその他の進行性固形がん患者を対象にCDK4阻害剤BGB-4... \n", "201 遺伝子HSD17B4高メチル化を有するHER2陽性ER陰性乳癌における非手術療法の有用性を評... \n", "234 DNAミスマッチ修復機構欠損を有する進行性・転移性子宮体がん患者を対象に、一次治療として化学... \n", "257 免疫チェックポイント阻害剤に対してacquired resistanceとなったがん患者を対... \n", "271 治療歴のある HER2 陽性又は HER2 低発現の切除不能又は転移性乳癌の被験者を対象に，... \n", "279 未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、... \n", "311 Programmed death-ligand（PD-L1）陽性の局所再発手術不能又は転移性... \n", "396 根治的局所治療（化学療法の併用または非併用）を受けて疾患の兆候のない、再発リスクが中間～高リ... \n", "467 [M23-647]B細胞性悪性腫瘍患者を対象としたBTK 分解誘導化合物ABBV-101 の... \n", "509 EMBER-4：2～5年間の術後内分泌療法による前治療歴を有する再発高リスクのER+、HER... \n", "514 MELK阻害剤OTS167POにおける転移性・進行性乳がん患者を対象とした安全性、忍容性およ... \n", "515 根治的局所領域療法（化学療法の併用または非併用）および標準補助内分泌療法（ET）を少なくとも... \n", "582 切除不能進行再発食道がん及び胃がんに対する免疫チェックポイント阻害剤(ICI)と光線力学的療... \n", "610 術前薬物療法後の外科的切除時に乳房及び／又は腋窩リンパ節に浸潤性残存病変を有するステージI～... \n", "693 PIK3CA H1047R変異を有する進行乳がん患者及びその他の固形がん患者を対象としたLO... \n", "708 初回腫瘍減量手術肉眼的完全切除後の進行卵巣癌を対象にパクリタキセル、カルボプラチン投与後のニ... \n", "753 局所進行（切除不能）又は転移性のホルモン受容体陽性／ヒト上皮細胞成長因子受容体2 陰性乳癌を... \n", "802 進行又は転移性固形がん患者を対象としたAZD8205の単独療法及び他の抗がん剤との併用療法に... \n", "865 乳がんを有する成人女性及び健康成人女性を対象とした乳房用マイクロ波画像診断装置IGS-000... \n", "\n", " Target \\\n", "12 肺がん \n", "59 ・進行性固形腫瘍・進行性乳がん・転移性乳がん・ホルモン受容体陽性乳がん・ホルモン受容体陽性乳... \n", "201 乳がん \n", "234 子宮体がん \n", "257 肺がん、食道がん、胃癌等の固形がん \n", "271 HER2 陽性又は HER2 低発現の切除不能又は転移性乳癌 \n", "279 乳癌 \n", "311 乳癌 \n", "396 乳がん、早期乳がん \n", "467 血液がん \n", "509 乳癌 \n", "514 再発・難治性の局所進行性・転移性乳がん及びトリプルネガティブ乳がん \n", "515 乳がん、早期乳がん \n", "582 食道がん、胃がん \n", "610 乳癌 \n", "693 乳がん \n", "708 卵巣がん \n", "753 局所進行（切除不能）又は転移性乳癌 \n", "802 乳癌、胆道癌、卵巣癌、子宮内膜癌 \n", "865 乳がん \n", "\n", " 研究・治験の目的 試験等のフェーズ 試験の種類 \\\n", "12 全身麻酔を要する肺切除術周術期における標準制吐療法であるグラニセトロン(Granisetro... 2 NaN \n", "59 サイクリン依存性キナーゼ 4 (CDK4) 阻害剤である BGB-43395 が、ホルモン受... 1 NaN \n", "201 HSD17B4高メチル化（HSD17B4 hypermethylation：HH）を有するH... 2 NaN \n", "234 dMMR/MSI-Hを有する再発又は進行子宮体癌患者を対象にdostarlimabの有効性及... 3 NaN \n", "257 免疫チェックポイント阻害剤に対してacquired resistanceとなったがん患者（非... 1-2 NaN \n", "271 治療 2 NaN \n", "279 未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、... 3 NaN \n", "311 PD-L1陽性の局所再発手術不能または転移性TNBC患者を対象に、デュルバルマブ併用または非... 3 NaN \n", "396 Treatment 3 NaN \n", "467 ・ABBV-101 の安全性を明らかにし，忍容性を確認する\\r\\n・ABBV-101 の経口... 1 NaN \n", "509 早期乳癌患者を対象としたimlunestrantと標準的な内分泌療法の比較試験 3 NaN \n", "514 再発・難治性の局所進行性または転移性乳がん患者に対して、OTS167を経口カプセルで投与する... 1 NaN \n", "515 Treatment 3 NaN \n", "582 一次治療として免疫チェックポイント阻害剤（ICI）を含む標準治療を予定される進行・再発食道が... 2 NaN \n", "610 術前薬物療法後の外科的切除時に乳房および／または腋窩リンパ節に浸潤性残存病変を有するI～II... 3 NaN \n", "693 LOXO-783 の単独投与及び他の抗がん剤との併用投与における 第 2 相試験の推奨用量、... 1 NaN \n", "708 診断時にステージIIIA/B/Cの高悪性度上皮性卵巣癌で、PDS後に残存病変を認めない患者を... 2 NaN \n", "753 Treatment 3 NaN \n", "802 進行又は転移性固形がんに対する治療法候補としての新規化合物AZD8205の研究 1-2 NaN \n", "865 本治験の目的は，乳がん又はその疑いのある者，並びに乳がん又はその疑いがない者を対象に乳房画像... 2 NaN \n", "\n", " 無作為化 盲検化 ... purpose \\\n", "12 無作為化比較 非盲検 ... treatment purpose \n", "59 非無作為化比較 非盲検 ... treatment purpose \n", "201 単一群 非盲検 ... diagnostic purpose \n", "234 無作為化比較 非盲検 ... treatment purpose \n", "257 単一群 非盲検 ... treatment purpose \n", "271 無作為化比較 非盲検 ... treatment purpose \n", "279 無作為化比較 非盲検 ... treatment purpose \n", "311 無作為化比較 非盲検 ... treatment purpose \n", "396 無作為化比較 非盲検 ... treatment purpose \n", "467 単一群 非盲検 ... treatment purpose \n", "509 無作為化比較 非盲検 ... treatment purpose \n", "514 単一群 非盲検 ... treatment purpose \n", "515 無作為化比較 非盲検 ... treatment purpose \n", "582 単一群 非盲検 ... treatment purpose \n", "610 無作為化比較 非盲検 ... treatment purpose \n", "693 単一群 非盲検 ... treatment purpose \n", "708 無作為化比較 非盲検 ... treatment purpose \n", "753 無作為化比較 二重盲検 ... treatment purpose \n", "802 単一群 非盲検 ... treatment purpose \n", "865 単一群 単盲検 ... screening \n", "\n", " Inclusion Criteria \\\n", "12 (1) Patients scheduled to undergo anatomical l... \n", "59 -Phase 1a (Dose Escalation) and 1b (Dose Expan... \n", "201 1. Histologically confirmed invasive breast ca... \n", "234 1. Female patient is at least 18 years of age,... \n", "257 1. Japanese patients aged 20 years or older at... \n", "271 (1) Male or female, aged >=18 years at the tim... \n", "279 1. Participant must be >= 18 years at the time... \n", "311 Histologically or cytologically documented loc... \n", "396 - Women and Men; 18 years or more at the time ... \n", "467 -For Dose Escalation (Part 1) only: Participan... \n", "509 -Have a diagnosis of ER+, HER2- early-stage, r... \n", "514 Dose Escalation and Dose Expansion Cohorts\\r\\n... \n", "515 - Women and Men, greater than or equal to 18 y... \n", "582 The patient with unresectable or recurrent eso... \n", "610 Participant must be >= 18 years at the time of... \n", "693 -Have advanced breast cancer or another solid ... \n", "708 1. Female patient >= 18 years of age.\\r\\n2. Si... \n", "753 Key inclusion criteria for both phases:\\n\\r\\n1... \n", "802 - Age18 years or more\\n\\r\\n- Relapsed/metastat... \n", "865 Breast cancer (including suspected) \\r\\n(1) On... \n", "\n", " Exclusion Criteria Age Minimum \\\n", "12 (1) Patients with a history of allergy to the ... 18age old over \n", "59 -Prior therapy selectively targeting CDK4 (pri... 18age old over \n", "201 1. History of other malignancy within the last... 20age old over \n", "234 1. Patient has received neoadjuvant/adjuvant s... 18age old over \n", "257 1. Patients with a history of severe drug hype... 20age old over \n", "271 (1) Presence of brain or subdural metastases, ... 18age old over \n", "279 1. As judged by the investigator, any evidence... 18age old over \n", "311 As judged by investigator, severe or uncontrol... 18age old over \n", "396 - Inoperable locally advanced or metastatic br... 18age old over \n", "467 -Previously treated with a Bruton's tyrosine k... 18age old over \n", "509 -Have any evidence of metastatic disease (incl... 18age old over \n", "514 Dose Escalation and Expansion Cohorts\\r\\n1.\\tW... 18age old over \n", "515 - Inoperable locally advanced or metastatic br... 18age old over \n", "582 Has been diagnosed with HER2-positive gastric ... 18age old over \n", "610 Stage IV (metastatic) TNBC.\\r\\nHistory of prio... 18age old over \n", "693 -Medical Conditions\\r\\n -Colorectal cancer\\r\\n... 18age old over \n", "708 1. Patient with clear cell adenocarcinoma or c... 18age old over \n", "753 Key exclusion criteria for both phases:\\r\\n1. ... 18age old over \n", "802 - Treatment with any of the following:\\n\\r\\n1.... 18age old over \n", "865 (1) There is trauma with bleeding on the breas... 20age old over \n", "\n", " Age Maximum Gender Discontinuation Criteria \\\n", "12 No limit Both NaN \n", "59 No limit Both NaN \n", "201 No limit Female NaN \n", "234 No limit Female NaN \n", "257 No limit Both NaN \n", "271 No limit Both NaN \n", "279 No limit Both NaN \n", "311 No limit Both NaN \n", "396 No limit NaN NaN \n", "467 No limit Both NaN \n", "509 No limit Both NaN \n", "514 No limit Female NaN \n", "515 No limit Both NaN \n", "582 No limit Both NaN \n", "610 130age old under Both NaN \n", "693 No limit Both NaN \n", "708 No limit Female NaN \n", "753 99age old under Both NaN \n", "802 No limit Both NaN \n", "865 No limit Female NaN \n", "\n", " Keyword \\\n", "12 NaN \n", "59 NaN \n", "201 NaN \n", "234 NaN \n", "257 NaN \n", "271 NaN \n", "279 NaN \n", "311 NaN \n", "396 NaN \n", "467 NaN \n", "509 NaN \n", "514 NaN \n", "515 NaN \n", "582 Esophageal cancer, Gastric cancer \n", "610 NaN \n", "693 NaN \n", "708 NaN \n", "753 NaN \n", "802 NaN \n", "865 NaN \n", "\n", " Intervention(s) \\\n", "12 Treatment group\\r\\nAdminister GRA 1mg intraven... \n", "59 -Drug: BGB-43395\\r\\nPlanned doses administered... \n", "201 Omitting breast surgery after preoperative che... \n", "234 Patients will be randomized 1:1 to receive eit... \n", "257 Intramuscular injection of the investigational... \n", "271 Generic Name:NA\\r\\nStudy Treatment in dose opt... \n", "279 - Experimental arm: Dato-DXd plus durvalumab n... \n", "311 Arm 1: Dato-DXd + durvalumab\\r\\nArm 2: Investi... \n", "396 arm A: continue with SoC ET as directed by inv... \n", "467 Drug: ABBV-101\\r\\nOral: Tablet \n", "509 -Drug: Imlunestrant\\r\\n Administered orally.\\... \n", "514 This is a Phase I dose escalation/expansion mu... \n", "515 arm A: continue with SoC ET as directed by inv... \n", "582 Cohort A: Nivolumab(240mg, every 2 weeks or 36... \n", "610 Arm 1: Dato-DXd 6 mg/kg IV Q3W x 8 cycles + Du... \n", "693 -Drug: LOXO-783\\r\\n Oral\\r\\n Other Name: LY384... \n", "708 Patients will be included after surgery and wi... \n", "753 Capivasertib: PO BD 4 days on /3 days off per ... \n", "802 Sub-study1\\r\\nAZD8205 Monotherapy\\r\\nAZD8205 i... \n", "865 The test using IGS-0001 will be performed twic... \n", "\n", " TargetWord \n", "12 [肺がん] \n", "59 [進行性固形腫瘍, 進行性乳がん, 転移性乳がん, ホルモン受容体陽性乳がん, ホルモン受容... \n", "201 [乳がん] \n", "234 [子宮体がん] \n", "257 [肺がん, 食道がん, 胃癌等の固形がん] \n", "271 [HER2 陽性, HER2 低発現の切除不能, 転移性乳癌] \n", "279 [乳癌] \n", "311 [乳癌] \n", "396 [乳がん, 早期乳がん] \n", "467 [血液がん] \n", "509 [乳癌] \n", "514 [再発, 難治性の局所進行性, 転移性乳がん, トリプルネガティブ乳がん] \n", "515 [乳がん, 早期乳がん] \n", "582 [食道がん, 胃がん] \n", "610 [乳癌] \n", "693 [乳がん] \n", "708 [卵巣がん] \n", "753 [局所進行（切除不能）, 転移性乳癌] \n", "802 [乳癌, 胆道癌, 卵巣癌, 子宮内膜癌] \n", "865 [乳がん] \n", "\n", "[20 rows x 39 columns]" ] }, "execution_count": 68, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# ターゲットリスト全体を処理\n", "matched_indices = []\n", "target_vecs_list = []\n", "cosine_scores_list = []\n", "for idx, target_words in enumerate(basedf['TargetWord']):\n", " # ターゲット内の各単語をベクトル化\n", " target_vecs = encode_text(target_words)\n", " # コサイン類似度を計算\n", " cosine_scores = torch.nn.functional.cosine_similarity(query_vec, target_vecs)\n", " target_vecs_list.append(target_vecs)\n", " cosine_scores_list.append(cosine_scores)\n", " # 閾値を超えるか確認\n", " if (cosine_scores >= threshold).any(): # いずれかが閾値を超えている場合\n", " matched_indices.append(idx)\n", "\n", "# 抽出\n", "matched_df = basedf.iloc[matched_indices]\n", "matched_df" ] }, { "cell_type": "code", "execution_count": 69, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[tensor([0.8485]),\n", " tensor([0.8322, 0.8710]),\n", " tensor([0.8630]),\n", " tensor([0.8748, 0.8663]),\n", " tensor([0.9373]),\n", " tensor([0.8615, 0.8576]),\n", " tensor([0.8443]),\n", " tensor([0.8511]),\n", " tensor([0.8320, 0.8452]),\n", " tensor([0.8485]),\n", " tensor([0.8681]),\n", " tensor([0.8585]),\n", " tensor([0.8489]),\n", " tensor([0.8535]),\n", " tensor([0.9058]),\n", " tensor([0.8473, 0.9020]),\n", " tensor([0.8587]),\n", " tensor([0.8621]),\n", " tensor([0.8526]),\n", " tensor([0.8363]),\n", " tensor([0.8573]),\n", " tensor([0.7646, 0.7990]),\n", " tensor([0.9068, 0.8733, 0.8775]),\n", " tensor([0.8584]),\n", " tensor([0.8584]),\n", " tensor([0.8624]),\n", " tensor([0.8456]),\n", " tensor([0.8959]),\n", " tensor([0.8844]),\n", " tensor([0.8842, 0.9356, 0.9305, 0.9065, 0.8970]),\n", " tensor([0.8691, 0.8630, 0.8718]),\n", " tensor([0.8758]),\n", " tensor([0.8630]),\n", " tensor([0.8611]),\n", " tensor([0.8630]),\n", " tensor([0.8963]),\n", " tensor([0.8601]),\n", " tensor([0.8040, 0.8439]),\n", " tensor([0.8789, 0.8568]),\n", " tensor([0.8110, 0.8643, 0.8400, 0.9012]),\n", " tensor([0.8604]),\n", " tensor([0.8752, 0.8826, 0.8916, 0.8565]),\n", " tensor([0.9091, 0.8941, 0.8630]),\n", " tensor([0.8391, 0.8505]),\n", " tensor([0.8625]),\n", " tensor([0.8507]),\n", " tensor([0.8583, 0.8685]),\n", " tensor([0.8758]),\n", " tensor([0.8597]),\n", " tensor([0.8707]),\n", " tensor([0.8908]),\n", " tensor([0.8480]),\n", " tensor([0.8567]),\n", " tensor([0.8752, 0.8763, 0.8826]),\n", " tensor([0.8630]),\n", " tensor([0.8497]),\n", " tensor([0.8395, 0.8532, 0.7941, 0.8583]),\n", " tensor([0.8630]),\n", " tensor([0.8254, 0.8873]),\n", " tensor([0.8758]),\n", " tensor([0.8806]),\n", " tensor([0.8152, 0.8277, 0.8346, 0.8746]),\n", " tensor([0.8478]),\n", " tensor([0.8758]),\n", " tensor([0.8781]),\n", " tensor([0.8615]),\n", " tensor([0.8446, 0.8641]),\n", " tensor([0.8733]),\n", " tensor([0.8507]),\n", " tensor([0.8584]),\n", " tensor([0.8630]),\n", " tensor([0.8545]),\n", " tensor([0.8819]),\n", " tensor([0.8969, 0.8613, 0.8482, 0.8693, 0.8598]),\n", " tensor([0.8688]),\n", " tensor([0.8625]),\n", " tensor([0.8455, 0.8619]),\n", " tensor([0.8788]),\n", " tensor([0.8540]),\n", " tensor([0.8625]),\n", " tensor([0.8630]),\n", " tensor([0.8664]),\n", " tensor([0.8630]),\n", " tensor([0.8459, 0.8211, 0.8484]),\n", " tensor([0.8508, 0.8430, 0.8313, 0.8702]),\n", " tensor([0.8469]),\n", " tensor([0.9016]),\n", " tensor([0.8630]),\n", " tensor([0.8379]),\n", " tensor([0.8558, 0.8561, 0.8513]),\n", " tensor([0.8582, 0.8528]),\n", " tensor([0.8582]),\n", " tensor([0.8536]),\n", " tensor([0.8804, 0.8691, 0.8711]),\n", " tensor([0.8377, 0.8560, 0.8276]),\n", " tensor([0.8604]),\n", " tensor([0.8301, 0.8577]),\n", " tensor([0.8410]),\n", " tensor([0.8664]),\n", " tensor([0.8473, 0.8524, 0.8974]),\n", " tensor([0.8473, 0.8524, 0.8974]),\n", " tensor([1.]),\n", " tensor([0.8547]),\n", " tensor([0.8540]),\n", " tensor([0.8630]),\n", " tensor([0.8615, 0.8576, 0.8696]),\n", " tensor([0.9083, 0.8155, 0.8555, 0.8375, 0.8585]),\n", " tensor([0.8624, 0.8528]),\n", " tensor([0.8630]),\n", " tensor([0.8584]),\n", " tensor([0.8347]),\n", " tensor([0.8630]),\n", " tensor([0.8526]),\n", " tensor([0.9058]),\n", " tensor([0.8624]),\n", " tensor([0.8597]),\n", " tensor([0.8752, 0.8763, 0.8826]),\n", " tensor([0.8305, 0.8894]),\n", " tensor([0.9246]),\n", " tensor([0.8630]),\n", " tensor([0.8611]),\n", " tensor([0.9068, 0.8908]),\n", " tensor([0.8708]),\n", " tensor([0.8587]),\n", " tensor([0.8347]),\n", " tensor([0.8959]),\n", " tensor([0.8711]),\n", " tensor([0.8711]),\n", " tensor([0.8388, 0.8499]),\n", " tensor([0.9373, 0.9130, 0.8716]),\n", " tensor([0.8913, 0.8613, 0.8677, 0.8699, 0.8787, 0.8538, 0.8325]),\n", " tensor([0.8540]),\n", " tensor([0.8514]),\n", " tensor([0.8388, 0.8499]),\n", " tensor([0.8457, 0.8334, 0.9242]),\n", " tensor([0.8758]),\n", " tensor([0.8959]),\n", " tensor([0.8841]),\n", " tensor([0.9522]),\n", " tensor([0.8630]),\n", " tensor([0.8584]),\n", " tensor([0.8604]),\n", " tensor([0.8525]),\n", " tensor([0.8535]),\n", " tensor([0.8662]),\n", " tensor([0.8639]),\n", " tensor([0.8466, 0.8730]),\n", " tensor([0.8337, 0.9098]),\n", " tensor([0.8670]),\n", " tensor([0.8577]),\n", " tensor([0.8863, 0.8752]),\n", " tensor([0.8630]),\n", " tensor([0.8315, 0.8696]),\n", " tensor([0.8362, 0.9098]),\n", " tensor([0.8380]),\n", " tensor([0.8540]),\n", " tensor([0.9522]),\n", " tensor([0.8908]),\n", " tensor([0.8305, 0.8000, 0.8520]),\n", " tensor([0.8540]),\n", " tensor([0.8673]),\n", " tensor([0.8593]),\n", " tensor([0.8604]),\n", " tensor([0.8203, 0.8343, 0.7933]),\n", " tensor([0.8391, 0.8621]),\n", " tensor([0.8368, 0.8635]),\n", " tensor([0.8365, 0.8518, 0.8365, 0.8411]),\n", " tensor([0.8391, 0.8505]),\n", " tensor([0.8540]),\n", " tensor([0.8536, 0.8536]),\n", " tensor([0.8248, 0.7948, 0.8418, 0.8342, 0.8219]),\n", " tensor([0.8521]),\n", " tensor([0.8106, 0.8684]),\n", " tensor([0.8510]),\n", " tensor([0.8446]),\n", " tensor([0.8541, 0.8603]),\n", " tensor([0.8789, 0.8576, 0.8710]),\n", " tensor([0.8631]),\n", " tensor([0.8510]),\n", " tensor([0.8523]),\n", " tensor([0.8423]),\n", " tensor([0.8908]),\n", " tensor([0.8540]),\n", " tensor([0.8376, 0.8768]),\n", " tensor([0.8051, 0.8686]),\n", " tensor([0.8630]),\n", " tensor([0.8630, 0.8839]),\n", " tensor([0.8604]),\n", " tensor([0.8051]),\n", " tensor([0.8688]),\n", " tensor([0.8562]),\n", " tensor([0.8562, 0.8553, 0.8430, 0.8563]),\n", " tensor([0.8586]),\n", " tensor([0.8458, 0.8616]),\n", " tensor([0.8375, 0.8453]),\n", " tensor([0.8540]),\n", " tensor([0.8603, 0.8568, 0.8696]),\n", " tensor([0.8450]),\n", " tensor([0.8371, 0.8642, 0.8535]),\n", " tensor([0.8417, 0.8689]),\n", " tensor([0.9093]),\n", " tensor([0.8728]),\n", " tensor([0.9990, 0.9460]),\n", " tensor([0.8443, 0.8873]),\n", " tensor([0.8759]),\n", " tensor([0.8752]),\n", " tensor([0.8440]),\n", " tensor([0.7623, 0.8737, 0.8236, 0.8576, 0.8696, 0.8687]),\n", " tensor([0.8630]),\n", " tensor([0.8314, 0.8953]),\n", " tensor([0.8388, 0.8499]),\n", " tensor([0.8360, 0.8724]),\n", " tensor([0.8303, 0.8760, 0.8303, 0.8781, 0.8303, 0.8612]),\n", " tensor([0.8384, 0.8511]),\n", " tensor([0.8391, 0.8590]),\n", " tensor([0.8247, 0.8847, 0.8139, 0.8356]),\n", " tensor([0.8628]),\n", " tensor([0.8732, 0.8518, 0.7789]),\n", " tensor([0.8204, 0.8629]),\n", " tensor([0.8384, 0.7728, 0.8089, 0.8722]),\n", " tensor([0.8061, 0.8273]),\n", " tensor([0.8435]),\n", " tensor([0.8344, 0.8426, 0.8477]),\n", " tensor([0.8838, 0.8694]),\n", " tensor([0.8535]),\n", " tensor([0.8411]),\n", " tensor([0.8804]),\n", " tensor([0.8485]),\n", " tensor([0.8155, 0.8625]),\n", " tensor([0.8511, 0.8162, 0.8280, 0.8599, 0.8879]),\n", " tensor([0.8422, 0.7904, 0.8108, 0.8225]),\n", " tensor([0.8901]),\n", " tensor([0.8508]),\n", " tensor([0.8806, 0.8688]),\n", " tensor([0.8786]),\n", " tensor([0.9182]),\n", " tensor([0.8405]),\n", " tensor([0.8563]),\n", " tensor([0.8630]),\n", " tensor([0.8495]),\n", " tensor([0.8110, 0.8427, 0.8705]),\n", " tensor([0.8525]),\n", " tensor([0.8106, 0.8752]),\n", " tensor([0.8688]),\n", " tensor([0.8604]),\n", " tensor([0.8540]),\n", " tensor([0.8540]),\n", " tensor([0.8809]),\n", " tensor([0.8789, 0.8576, 0.8790, 0.8781]),\n", " tensor([0.8525]),\n", " tensor([0.8846]),\n", " tensor([0.9068]),\n", " tensor([0.8602, 0.8571, 0.8578, 0.8484]),\n", " tensor([0.8347]),\n", " tensor([0.8501, 0.9079]),\n", " tensor([0.9522]),\n", " tensor([0.8107, 0.8577]),\n", " tensor([0.8938, 0.8839, 0.8568]),\n", " tensor([0.8391, 0.8348, 0.9305, 0.9071]),\n", " tensor([0.9990, 0.9460]),\n", " tensor([0.8587, 0.8379, 0.8649]),\n", " tensor([0.8540]),\n", " tensor([0.8604]),\n", " tensor([0.8264, 0.8635, 0.8576]),\n", " tensor([0.8264]),\n", " tensor([0.9071]),\n", " tensor([0.8395]),\n", " tensor([0.8179, 0.8403]),\n", " tensor([0.8630]),\n", " tensor([0.8540]),\n", " tensor([0.8409, 0.8956]),\n", " tensor([0.8602, 0.8630, 0.8836]),\n", " tensor([0.8908]),\n", " tensor([0.9068]),\n", " tensor([0.8309, 0.8598]),\n", " tensor([0.8576, 0.8500]),\n", " tensor([0.8625]),\n", " tensor([0.8804]),\n", " tensor([0.8359, 0.8612, 0.8127]),\n", " tensor([0.8163, 0.8752]),\n", " tensor([0.8806]),\n", " tensor([0.9118, 0.9205]),\n", " tensor([0.8786]),\n", " tensor([0.8577]),\n", " tensor([0.8506]),\n", " tensor([0.8201, 0.8684]),\n", " tensor([0.8499]),\n", " tensor([0.8391, 0.8505]),\n", " tensor([0.8604]),\n", " tensor([0.8535]),\n", " tensor([0.9522]),\n", " tensor([0.8773]),\n", " tensor([0.8280]),\n", " tensor([0.8488]),\n", " tensor([0.8540]),\n", " tensor([0.8459]),\n", " tensor([0.8514, 0.8533]),\n", " tensor([0.8249, 0.8579]),\n", " tensor([0.8510]),\n", " tensor([0.8597, 0.8345, 0.9077, 0.8147, 0.8507]),\n", " tensor([0.8188, 0.8284]),\n", " tensor([0.8393]),\n", " tensor([0.8322, 0.8710]),\n", " tensor([0.8630]),\n", " tensor([0.8823]),\n", " tensor([0.8659]),\n", " tensor([0.8630]),\n", " tensor([0.8630]),\n", " tensor([0.8907, 0.8770, 0.8625, 0.9053, 0.8711, 0.8543, 0.8684]),\n", " tensor([0.8658]),\n", " tensor([0.8503, 0.8373]),\n", " tensor([0.8517]),\n", " tensor([0.8630]),\n", " tensor([0.8540]),\n", " tensor([0.8365, 0.7996, 0.8437, 0.8365, 0.7996, 0.8523]),\n", " tensor([0.8804]),\n", " tensor([0.8908]),\n", " tensor([0.8604]),\n", " tensor([0.8649, 0.8198, 0.8309]),\n", " tensor([0.8733]),\n", " tensor([1.]),\n", " tensor([0.8749]),\n", " tensor([0.8733, 0.8395, 0.8583]),\n", " tensor([0.8806]),\n", " tensor([0.9128]),\n", " tensor([0.8630]),\n", " tensor([0.8540]),\n", " tensor([0.8552]),\n", " tensor([0.8908]),\n", " tensor([0.8577]),\n", " tensor([0.8604]),\n", " tensor([0.8463, 0.8154, 0.8398, 0.8542, 0.8673, 0.8961]),\n", " tensor([0.8551, 0.9088]),\n", " tensor([0.8841]),\n", " tensor([0.8358]),\n", " tensor([0.8806]),\n", " tensor([0.8789, 0.8576]),\n", " tensor([0.8630]),\n", " tensor([0.8232, 0.9240]),\n", " tensor([0.8584]),\n", " tensor([0.8242]),\n", " tensor([0.8471]),\n", " tensor([0.8499]),\n", " tensor([0.8526]),\n", " tensor([0.8513, 0.8181, 0.8358, 0.8695]),\n", " tensor([0.8433]),\n", " tensor([0.8630]),\n", " tensor([0.8789, 0.8576]),\n", " tensor([0.8405]),\n", " tensor([0.8584]),\n", " tensor([0.8580]),\n", " tensor([0.8496]),\n", " tensor([0.8540]),\n", " tensor([0.8434, 0.8146, 0.8664]),\n", " tensor([0.8375, 0.8562]),\n", " tensor([0.8319, 0.8758]),\n", " tensor([0.8725, 0.8489, 0.8651]),\n", " tensor([0.9505, 0.8707, 0.9088, 0.8914]),\n", " tensor([0.8604]),\n", " tensor([0.8315, 0.8696]),\n", " tensor([0.8535]),\n", " tensor([0.8290, 0.8370]),\n", " tensor([0.8908]),\n", " tensor([0.8182, 0.8864, 0.8526]),\n", " tensor([0.8841]),\n", " tensor([0.8685]),\n", " tensor([0.8106, 0.8752]),\n", " tensor([0.8770]),\n", " tensor([0.8430, 0.8630, 0.8430, 0.8706, 0.8430, 0.8658]),\n", " tensor([0.8265]),\n", " tensor([0.8526]),\n", " tensor([0.8559]),\n", " tensor([0.8161, 0.8664]),\n", " tensor([0.8586]),\n", " tensor([1.]),\n", " tensor([0.8256, 0.8344]),\n", " tensor([0.8712, 0.8568]),\n", " tensor([0.8467]),\n", " tensor([0.8604]),\n", " tensor([0.8669, 0.8661]),\n", " tensor([0.8630, 0.8427, 0.8530, 0.8497, 0.8509, 0.8302]),\n", " tensor([0.8336, 0.8468, 0.8465]),\n", " tensor([0.8380, 0.8434, 0.8507]),\n", " tensor([0.8220, 0.8638]),\n", " tensor([0.8623]),\n", " tensor([0.8540]),\n", " tensor([0.8758]),\n", " tensor([0.8375, 0.8453]),\n", " tensor([0.8776]),\n", " tensor([0.8584]),\n", " tensor([0.8583, 0.8142]),\n", " tensor([0.7930, 0.7781, 0.8385]),\n", " tensor([0.8479]),\n", " tensor([0.8923, 0.8914, 0.7692, 0.8157]),\n", " tensor([0.8373, 0.8890]),\n", " tensor([0.8785]),\n", " tensor([0.8584]),\n", " tensor([0.8405]),\n", " tensor([0.8159, 0.8669]),\n", " tensor([0.8381, 0.8489]),\n", " tensor([0.8450]),\n", " tensor([0.8347]),\n", " tensor([0.8554]),\n", " tensor([0.8535, 0.8582]),\n", " tensor([0.8536]),\n", " tensor([0.8561]),\n", " tensor([0.8540]),\n", " tensor([0.8914]),\n", " tensor([0.8101, 0.8767, 0.8590, 0.8398]),\n", " tensor([0.8797]),\n", " tensor([0.8584]),\n", " tensor([0.8615, 0.8568, 0.8696]),\n", " tensor([0.8419]),\n", " tensor([0.8385]),\n", " tensor([0.8662]),\n", " tensor([0.8648])]" ] }, "execution_count": 69, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cosine_scores_list" ] }, { "cell_type": "code", "execution_count": 70, "metadata": {}, "outputs": [], "source": [ "# 全データのターゲット列をベクトル化\n", "target_list = basedf['Target'].tolist()\n", "\n", "target_vecs = encode_text(target_list)\n", " # コサイン類似度を計算\n", "cosine_scores = torch.nn.functional.cosine_similarity(query_vec, target_vecs)" ] }, { "cell_type": "code", "execution_count": 71, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
JRCT IDNCT NoJapicCTI NoTitleTarget研究・治験の目的試験等のフェーズ試験の種類無作為化盲検化...purposeInclusion CriteriaExclusion CriteriaAge MinimumAge MaximumGenderDiscontinuation CriteriaKeywordIntervention(s)TargetWord
12jRCT2041240130NaNNaN中等度-高度PONV発症リスクを有する肺切除術予定患者の周術期管理におけるGranisetr...肺がん全身麻酔を要する肺切除術周術期における標準制吐療法であるグラニセトロン(Granisetro...2NaN無作為化比較非盲検...treatment purpose(1) Patients scheduled to undergo anatomical l...(1) Patients with a history of allergy to the ...18age old overNo limitBothNaNNaNTreatment group\\r\\nAdminister GRA 1mg intraven...[肺がん]
201jRCT2052240059NaNNaN遺伝子HSD17B4高メチル化を有するHER2陽性ER陰性乳癌における非手術療法の有用性を評...乳がんHSD17B4高メチル化（HSD17B4 hypermethylation：HH）を有するH...2NaN単一群非盲検...diagnostic purpose1. Histologically confirmed invasive breast ca...1. History of other malignancy within the last...20age old overNo limitFemaleNaNNaNOmitting breast surgery after preoperative che...[乳がん]
279jRCT2031230723NCT06112379NaN未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、...乳癌未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、...3NaN無作為化比較非盲検...treatment purpose1. Participant must be >= 18 years at the time...1. As judged by the investigator, any evidence...18age old overNo limitBothNaNNaN- Experimental arm: Dato-DXd plus durvalumab n...[乳癌]
311jRCT2061230102NCT06103864NaNProgrammed death-ligand（PD-L1）陽性の局所再発手術不能又は転移性...乳癌PD-L1陽性の局所再発手術不能または転移性TNBC患者を対象に、デュルバルマブ併用または非...3NaN無作為化比較非盲検...treatment purposeHistologically or cytologically documented loc...As judged by investigator, severe or uncontrol...18age old overNo limitBothNaNNaNArm 1: Dato-DXd + durvalumab\\r\\nArm 2: Investi...[乳癌]
396jRCT2061230074NCT05952557NaN根治的局所治療（化学療法の併用または非併用）を受けて疾患の兆候のない、再発リスクが中間～高リ...乳がん、早期乳がんTreatment3NaN無作為化比較非盲検...treatment purpose- Women and Men; 18 years or more at the time ...- Inoperable locally advanced or metastatic br...18age old overNo limitNaNNaNNaNarm A: continue with SoC ET as directed by inv...[乳がん, 早期乳がん]
509jRCT2031230109NCT05514054NaNEMBER-4：2～5年間の術後内分泌療法による前治療歴を有する再発高リスクのER+、HER...乳癌早期乳癌患者を対象としたimlunestrantと標準的な内分泌療法の比較試験3NaN無作為化比較非盲検...treatment purpose-Have a diagnosis of ER+, HER2- early-stage, r...-Have any evidence of metastatic disease (incl...18age old overNo limitBothNaNNaN-Drug: Imlunestrant\\r\\n Administered orally.\\...[乳癌]
515jRCT2031230096NCT05774951NaN根治的局所領域療法（化学療法の併用または非併用）および標準補助内分泌療法（ET）を少なくとも...乳がん、早期乳がんTreatment3NaN無作為化比較非盲検...treatment purpose- Women and Men, greater than or equal to 18 y...- Inoperable locally advanced or metastatic br...18age old overNo limitBothNaNNaNarm A: continue with SoC ET as directed by inv...[乳がん, 早期乳がん]
610jRCT2061220087NCT05629585NaN術前薬物療法後の外科的切除時に乳房及び／又は腋窩リンパ節に浸潤性残存病変を有するステージI～...乳癌術前薬物療法後の外科的切除時に乳房および／または腋窩リンパ節に浸潤性残存病変を有するI～II...3NaN無作為化比較非盲検...treatment purposeParticipant must be >= 18 years at the time of...Stage IV (metastatic) TNBC.\\r\\nHistory of prio...18age old over130age old underBothNaNNaNArm 1: Dato-DXd 6 mg/kg IV Q3W x 8 cycles + Du...[乳癌]
693jRCT2031220276NCT05307705NaNPIK3CA H1047R変異を有する進行乳がん患者及びその他の固形がん患者を対象としたLO...乳がんLOXO-783 の単独投与及び他の抗がん剤との併用投与における 第 2 相試験の推奨用量、...1NaN単一群非盲検...treatment purpose-Have advanced breast cancer or another solid ...-Medical Conditions\\r\\n -Colorectal cancer\\r\\n...18age old overNo limitBothNaNNaN-Drug: LOXO-783\\r\\n Oral\\r\\n Other Name: LY384...[乳がん]
865jRCT2052210099NaNNaN乳がんを有する成人女性及び健康成人女性を対象とした乳房用マイクロ波画像診断装置IGS-000...乳がん本治験の目的は，乳がん又はその疑いのある者，並びに乳がん又はその疑いがない者を対象に乳房画像...2NaN単一群単盲検...screeningBreast cancer (including suspected) \\r\\n(1) On...(1) There is trauma with bleeding on the breas...20age old overNo limitFemaleNaNNaNThe test using IGS-0001 will be performed twic...[乳がん]
\n", "

10 rows × 39 columns

\n", "
" ], "text/plain": [ " JRCT ID NCT No JapicCTI No \\\n", "12 jRCT2041240130 NaN NaN \n", "201 jRCT2052240059 NaN NaN \n", "279 jRCT2031230723 NCT06112379 NaN \n", "311 jRCT2061230102 NCT06103864 NaN \n", "396 jRCT2061230074 NCT05952557 NaN \n", "509 jRCT2031230109 NCT05514054 NaN \n", "515 jRCT2031230096 NCT05774951 NaN \n", "610 jRCT2061220087 NCT05629585 NaN \n", "693 jRCT2031220276 NCT05307705 NaN \n", "865 jRCT2052210099 NaN NaN \n", "\n", " Title Target \\\n", "12 中等度-高度PONV発症リスクを有する肺切除術予定患者の周術期管理におけるGranisetr... 肺がん \n", "201 遺伝子HSD17B4高メチル化を有するHER2陽性ER陰性乳癌における非手術療法の有用性を評... 乳がん \n", "279 未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、... 乳癌 \n", "311 Programmed death-ligand（PD-L1）陽性の局所再発手術不能又は転移性... 乳癌 \n", "396 根治的局所治療（化学療法の併用または非併用）を受けて疾患の兆候のない、再発リスクが中間～高リ... 乳がん、早期乳がん \n", "509 EMBER-4：2～5年間の術後内分泌療法による前治療歴を有する再発高リスクのER+、HER... 乳癌 \n", "515 根治的局所領域療法（化学療法の併用または非併用）および標準補助内分泌療法（ET）を少なくとも... 乳がん、早期乳がん \n", "610 術前薬物療法後の外科的切除時に乳房及び／又は腋窩リンパ節に浸潤性残存病変を有するステージI～... 乳癌 \n", "693 PIK3CA H1047R変異を有する進行乳がん患者及びその他の固形がん患者を対象としたLO... 乳がん \n", "865 乳がんを有する成人女性及び健康成人女性を対象とした乳房用マイクロ波画像診断装置IGS-000... 乳がん \n", "\n", " 研究・治験の目的 試験等のフェーズ 試験の種類 無作為化 \\\n", "12 全身麻酔を要する肺切除術周術期における標準制吐療法であるグラニセトロン(Granisetro... 2 NaN 無作為化比較 \n", "201 HSD17B4高メチル化（HSD17B4 hypermethylation：HH）を有するH... 2 NaN 単一群 \n", "279 未治療のトリプルネガティブ又はホルモン受容体低発現／HER2陰性乳癌の成人患者を対象として、... 3 NaN 無作為化比較 \n", "311 PD-L1陽性の局所再発手術不能または転移性TNBC患者を対象に、デュルバルマブ併用または非... 3 NaN 無作為化比較 \n", "396 Treatment 3 NaN 無作為化比較 \n", "509 早期乳癌患者を対象としたimlunestrantと標準的な内分泌療法の比較試験 3 NaN 無作為化比較 \n", "515 Treatment 3 NaN 無作為化比較 \n", "610 術前薬物療法後の外科的切除時に乳房および／または腋窩リンパ節に浸潤性残存病変を有するI～II... 3 NaN 無作為化比較 \n", "693 LOXO-783 の単独投与及び他の抗がん剤との併用投与における 第 2 相試験の推奨用量、... 1 NaN 単一群 \n", "865 本治験の目的は，乳がん又はその疑いのある者，並びに乳がん又はその疑いがない者を対象に乳房画像... 2 NaN 単一群 \n", "\n", " 盲検化 ... purpose \\\n", "12 非盲検 ... treatment purpose \n", "201 非盲検 ... diagnostic purpose \n", "279 非盲検 ... treatment purpose \n", "311 非盲検 ... treatment purpose \n", "396 非盲検 ... treatment purpose \n", "509 非盲検 ... treatment purpose \n", "515 非盲検 ... treatment purpose \n", "610 非盲検 ... treatment purpose \n", "693 非盲検 ... treatment purpose \n", "865 単盲検 ... screening \n", "\n", " Inclusion Criteria \\\n", "12 (1) Patients scheduled to undergo anatomical l... \n", "201 1. Histologically confirmed invasive breast ca... \n", "279 1. Participant must be >= 18 years at the time... \n", "311 Histologically or cytologically documented loc... \n", "396 - Women and Men; 18 years or more at the time ... \n", "509 -Have a diagnosis of ER+, HER2- early-stage, r... \n", "515 - Women and Men, greater than or equal to 18 y... \n", "610 Participant must be >= 18 years at the time of... \n", "693 -Have advanced breast cancer or another solid ... \n", "865 Breast cancer (including suspected) \\r\\n(1) On... \n", "\n", " Exclusion Criteria Age Minimum \\\n", "12 (1) Patients with a history of allergy to the ... 18age old over \n", "201 1. History of other malignancy within the last... 20age old over \n", "279 1. As judged by the investigator, any evidence... 18age old over \n", "311 As judged by investigator, severe or uncontrol... 18age old over \n", "396 - Inoperable locally advanced or metastatic br... 18age old over \n", "509 -Have any evidence of metastatic disease (incl... 18age old over \n", "515 - Inoperable locally advanced or metastatic br... 18age old over \n", "610 Stage IV (metastatic) TNBC.\\r\\nHistory of prio... 18age old over \n", "693 -Medical Conditions\\r\\n -Colorectal cancer\\r\\n... 18age old over \n", "865 (1) There is trauma with bleeding on the breas... 20age old over \n", "\n", " Age Maximum Gender Discontinuation Criteria Keyword \\\n", "12 No limit Both NaN NaN \n", "201 No limit Female NaN NaN \n", "279 No limit Both NaN NaN \n", "311 No limit Both NaN NaN \n", "396 No limit NaN NaN NaN \n", "509 No limit Both NaN NaN \n", "515 No limit Both NaN NaN \n", "610 130age old under Both NaN NaN \n", "693 No limit Both NaN NaN \n", "865 No limit Female NaN NaN \n", "\n", " Intervention(s) TargetWord \n", "12 Treatment group\\r\\nAdminister GRA 1mg intraven... [肺がん] \n", "201 Omitting breast surgery after preoperative che... [乳がん] \n", "279 - Experimental arm: Dato-DXd plus durvalumab n... [乳癌] \n", "311 Arm 1: Dato-DXd + durvalumab\\r\\nArm 2: Investi... [乳癌] \n", "396 arm A: continue with SoC ET as directed by inv... [乳がん, 早期乳がん] \n", "509 -Drug: Imlunestrant\\r\\n Administered orally.\\... [乳癌] \n", "515 arm A: continue with SoC ET as directed by inv... [乳がん, 早期乳がん] \n", "610 Arm 1: Dato-DXd 6 mg/kg IV Q3W x 8 cycles + Du... [乳癌] \n", "693 -Drug: LOXO-783\\r\\n Oral\\r\\n Other Name: LY384... [乳がん] \n", "865 The test using IGS-0001 will be performed twic... [乳がん] \n", "\n", "[10 rows x 39 columns]" ] }, "execution_count": 71, "metadata": {}, "output_type": "execute_result" } ], "source": [ "matched_indices_d = (cosine_scores >= threshold).nonzero().tolist()\n", "# 入れ子リストをフラットなリストに変換\n", "flat_indices_d = [idx[0] for idx in matched_indices_d]\n", "\n", "# 抽出\n", "matched_df_d = basedf.iloc[flat_indices_d]\n", "matched_df_d\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n" ] } ], "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 }