DeconDTN: Deconfounding Deep Transformer Networks for Clinical NLP
DeconDTN:为临床 NLP 解构深度 Transformer 网络
基本信息
- 批准号:10711315
- 负责人:
- 金额:$ 31.12万
- 依托单位:
- 依托单位国家:美国
- 项目类别:
- 财政年份:2022
- 资助国家:美国
- 起止时间:2022-06-01 至 2026-02-28
- 项目状态:未结题
- 来源:
- 关键词:AccelerationAdverse effectsAffectAfrican AmericanAfrican American populationAlzheimer disease detectionAlzheimer&aposs DiseaseAlzheimer&aposs disease diagnosisAlzheimer&aposs disease patientAwardCaregiversCaringClassificationClinicalCognitiveConfounding Factors (Epidemiology)DataData ProvenanceData SetDementiaDetectionDiagnosisDiseaseEarly DiagnosisElderlyEmotionalEquityEthnic OriginEthnic PopulationFamilyFamily RelationshipFinancial HardshipFutureGenetic TranscriptionHealth Care CostsHealthcare SystemsHispanic AmericansIndividualInequityInstitutionLanguageLearningLinguisticsMeasuresMediatingMethodsModelingModificationMonitorNatural Language ProcessingNatureNeural Network SimulationNeurobehavioral ManifestationsOutcomeParentsParticipantPatientsPerformancePersonalityPersonsPopulationPrognosisRaceResearchSamplingSemanticsSeriesSiteSocial isolationSocietiesSpeechStructureSymptomsTextTimeTrainingTranscriptUncertaintyUnderrepresented PopulationsVariantWorkautomated speech recognitioncognitive changecognitive functioncognitive taskcostcost effectivedeep learningdeep learning modeldetection methodethnic disparityimprovedmachine learning classifiermachine learning modelmarginalizationneural network architectureneural network classifierracial disparityracial populationresponsesoundtool
项目摘要
Abstract
This proposal relates to ongoing efforts to develop automated methods for the detection of linguistic
manifestations of cognitive changes in Alzheimer’s Disease (AD). These methods have the potential to alleviate
the personal and societal burden of AD, by reducing time to diagnosis. Delayed AD diagnosis has adverse effects
on care planning and family relationships, and has been estimated to cost the healthcare system close to $8
trillion dollars. Language reflects cognitive status, and contemporary neural network models have been shown
to discriminate between transcribed speech from patients with AD and that from healthy controls with promising
accuracy. However, most of this work has been conducted in the context of recorded responses to picture
description tasks, which are not suitable for repeated, continuous or passive monitoring for linguistic indicators
of AD-related decline. In contrast, our recent work has identified the task-agnostic linguistic construct of semantic
coherence as a sound basis for machine learning models for detection of language from people with AD in casual
conversations, with classification performance in this context exceeding that of Bidirectional Encoder
Representations from Transformers (BERT) based classifiers, the state-of-the-art for text-based AD detection.
Unfortunately, recent work has shown that automated coherence estimates are vulnerable to bias, with lower
estimates for speech from people identifying as black irrespective of diagnosis. Interrogation of coherence- and
BERT-based classifiers reveals they dramatically underperform in this group also. For the parent award for this
supplement proposal (R01 LM0104056), we will develop methods to debias deep transformer networks to
mitigate for the confounding variable of provenance in multi-institutional datasets, culminating in the release of
a toolkit to Deconfound Deep Transformer Networks, the DeconDTN suite. In this proposal we will apply these
methods to deconfound coherence- and BERT-based AD detection models for the confounding variable of race
and/or ethnicity, and evaluate the effects on model performance across groups. In addition, we will evaluate the
utility of fine-tuning semantic coherence models and models used to generate automated transcripts on text and
speech from the recently-released Corpus of Regional African American Language (CORAAL). We hypothesize
that both improvements in these underlying models, and explicit deconfounding for race/ethnicity will reduce the
performance differential across groups, resulting in equitable models for language-based AD detection.
摘要
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Trevor Cohen其他文献
Trevor Cohen的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Trevor Cohen', 18)}}的其他基金
DeconDTN: Deconfounding Deep Transformer Networks for Clinical NLP
DeconDTN:为临床 NLP 解构深度 Transformer 网络
- 批准号:
10626888 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Professional to Plain Language Neural Translation: A Path Toward Actionable Health Information
专业到通俗语言的神经翻译:通向可行健康信息的道路
- 批准号:
10349319 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Professional to Plain Language Neural Translation: A Path Toward Actionable Health Information
专业到通俗语言的神经翻译:通向可行健康信息的道路
- 批准号:
10579898 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
DeconDTN: Deconfounding Deep Transformer Networks for Clinical NLP
DeconDTN:为临床 NLP 解构深度 Transformer 网络
- 批准号:
10467107 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Computerized assessment of linguistic indicators of lucidity in Alzheimer's Disease dementia
阿尔茨海默病痴呆症语言清醒度指标的计算机化评估
- 批准号:
10093304 - 财政年份:2020
- 资助金额:
$ 31.12万 - 项目类别:
Using Biomedical Knowledge to Identify Plausible Signals for Pharmacovigilance
利用生物医学知识识别药物警戒的合理信号
- 批准号:
8914098 - 财政年份:2013
- 资助金额:
$ 31.12万 - 项目类别:
Using Biomedical Knowledge to Identify Plausible Signals for Pharmacovigilance
利用生物医学知识识别药物警戒的合理信号
- 批准号:
8727094 - 财政年份:2013
- 资助金额:
$ 31.12万 - 项目类别:
Encoding Semantic Knowledge in Vector Space for Biomedical Information
在生物医学信息的向量空间中编码语义知识
- 批准号:
8138564 - 财政年份:2010
- 资助金额:
$ 31.12万 - 项目类别:
Encoding Semantic Knowledge in Vector Space for Biomedical Information
在生物医学信息的向量空间中编码语义知识
- 批准号:
7977263 - 财政年份:2010
- 资助金额:
$ 31.12万 - 项目类别:
相似海外基金
Unraveling Adverse Effects of Checkpoint Inhibitors Using iPSC-derived Cardiac Organoids
使用 iPSC 衍生的心脏类器官揭示检查点抑制剂的副作用
- 批准号:
10591918 - 财政年份:2023
- 资助金额:
$ 31.12万 - 项目类别:
Optimization of mRNA-LNP vaccine for attenuating adverse effects and analysis of mechanism behind adverse effects
mRNA-LNP疫苗减轻不良反应的优化及不良反应机制分析
- 批准号:
23K15383 - 财政年份:2023
- 资助金额:
$ 31.12万 - 项目类别:
Grant-in-Aid for Early-Career Scientists
Elucidation of adverse effects of combined exposure to low-dose chemicals in the living environment on allergic diseases and attempts to reduce allergy
阐明生活环境中低剂量化学品联合暴露对过敏性疾病的不良影响并尝试减少过敏
- 批准号:
23H03556 - 财政年份:2023
- 资助金额:
$ 31.12万 - 项目类别:
Grant-in-Aid for Scientific Research (B)
Green tea-based nano-enhancer as an adjuvant for amplified efficacy and reduced adverse effects in anti-angiogenic drug treatments
基于绿茶的纳米增强剂作为抗血管生成药物治疗中增强疗效并减少不良反应的佐剂
- 批准号:
23K17212 - 财政年份:2023
- 资助金额:
$ 31.12万 - 项目类别:
Grant-in-Aid for Early-Career Scientists
Effects of Tobacco Heating System on the male reproductive function and towards to the reduce of the adverse effects.
烟草加热系统对男性生殖功能的影响以及减少不利影响。
- 批准号:
22H03519 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Grant-in-Aid for Scientific Research (B)
Mitigating the Adverse Effects of Ultrafines in Pressure Filtration of Oil Sands Tailings
减轻油砂尾矿压力过滤中超细粉的不利影响
- 批准号:
563657-2021 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Alliance Grants
1/4-Deciphering Mechanisms of ECT Outcomes and Adverse Effects (DECODE)
1/4-破译ECT结果和不良反应的机制(DECODE)
- 批准号:
10521849 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
4/4-Deciphering Mechanisms of ECT Outcomes and Adverse Effects (DECODE)
4/4-破译ECT结果和不良反应的机制(DECODE)
- 批准号:
10671022 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
2/4 Deciphering Mechanisms of ECT Outcomes and Adverse Effects (DECODE)
2/4 ECT 结果和不良反应的破译机制(DECODE)
- 批准号:
10670918 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Adverse Effects of Using Laser Diagnostics in High-Speed Compressible Flows
在高速可压缩流中使用激光诊断的不利影响
- 批准号:
RGPIN-2018-04753 - 财政年份:2022
- 资助金额:
$ 31.12万 - 项目类别:
Discovery Grants Program - Individual