Isabelle Lee

ORCID: 0000-0003-4125-2610
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Contact Dermatitis and Allergies
  • Animal testing and alternatives
  • Intelligent Tutoring Systems and Adaptive Learning
  • Pesticide Exposure and Toxicity
  • Topic Modeling
  • Natural Language Processing Techniques
  • Allergic Rhinitis and Sensitization
  • Antimicrobial agents and applications
  • Advanced Text Analysis Techniques
  • Computability, Logic, AI Algorithms
  • AI-based Problem Solving and Planning
  • Peroxisome Proliferator-Activated Receptors
  • Biomedical Ethics and Regulation
  • Biochemical and biochemical processes
  • Wnt/β-catenin signaling in development and cancer
  • Cancer, Lipids, and Metabolism
  • Estrogen and related hormone effects
  • Drug-Induced Adverse Reactions
  • Pharmaceutical Economics and Policy
  • Text and Document Classification Technologies
  • Dermatology and Skin Diseases
  • Neural Networks and Applications
  • Pharmacological Effects of Medicinal Plants
  • Acne and Rosacea Treatments and Effects
  • Constraint Satisfaction and Optimization

Research Institute for Fragrance Materials
2021-2025

Translational Therapeutics (United States)
2020

University of Pennsylvania
2018

Gladstone Institutes
1998

Cardiovascular Institute Hospital
1998

Osaka City University
1998

University of California, San Francisco
1998

The commitment to develop a roadmap for phasing out the use of animals chemical safety assessments was part European Commission's response Citizens' Initiative "Save Cruelty-Free Cosmetics - Commit Europe Without Animal Testing". aims outline milestones and specific actions be implemented in short long-term ultimately phase animal testing assessments. To advance this goal help define structure roadmap, multi-stakeholder roundtable workshop organised by five protection non-governmental...

10.14573/altex.2503241 article EN cc-by ALTEX 2025-01-01

Development of New Approach Methodologies (NAMs) capable providing a No Expected Sensitization Induction Level (NESIL) value remains high priority for the fragrance industry conducting Quantitative Risk Assesment (QRA) to evaluate dermal sensitization. The in vitro GARDskin assay was recently adopted by OECD (TG 442E) hazard identification skin sensitizers. Continuous potency predictions are derived using modified protocol that incorporates dose-response measurements. Linear regression...

10.1016/j.yrtph.2024.105597 article EN cc-by Regulatory Toxicology and Pharmacology 2024-03-08

Reliable human potency data are necessary for conducting quantitative risk assessments, as well development and validation of new nonanimal methods skin sensitization assessments. Previously, fragrance materials was derived primarily from or the local lymph node assay.This study aimed to define via weight evidence approach, incorporating all available human, animal, in vitro, chemico, silico data.All on 106 were considered assign each material into 1 6 defined categories (extreme, strong,...

10.1097/der.0000000000000854 article EN cc-by-nc-nd Dermatitis 2022-02-16

Abstract —Hepatic expression of apolipoprotein (apo) B mRNA–editing enzyme catalytic polypeptide 1 (APOBEC-1) has been proposed as a gene therapy approach for lowering plasma low density lipoprotein (LDL) levels. However, high-level APOBEC-1 in transgenic mouse and rabbit livers causes liver dysplasia hepatocellular carcinoma. To determine the physiological pathological effects low-level hepatic APOBEC-1, we used 52-kb rat genomic clone (RE4) to generate mice expressing levels (2 5 times...

10.1161/01.atv.18.6.1013 article EN Arteriosclerosis Thrombosis and Vascular Biology 1998-06-01

Some fragrance ingredients may have the potential to induce skin sensitization in humans but can still be safely formulated into consumer products. Quantitative Risk Assessment (QRA) for dermal is required determine safe levels at which sensitizers incorporated The no expected induction level or NESIL point of departure QRA. Sensitization assessment factors are applied acceptable exposure would general population. This paper details key steps involved deriving a weight evidence (WoE) given...

10.1016/j.fct.2021.112705 article EN cc-by Food and Chemical Toxicology 2021-11-24

The Research Institute for Fragrance Materials (RIFM) and Creme Global Cremeglobal.com partnered to develop an aggregate exposure model fragrance ingredients. provides a realistic estimate of the total ingredients individuals across population. Threshold Toxicological Concern (TTC) Dermal Sensitization (DST) were used demonstrate magnitude low materials. chronic systemic, inhalation, dermal 95th percentile exposures on approximately 3000 in RIFM's inventory compared their respective TTC or...

10.1016/j.yrtph.2024.105569 article EN cc-by Regulatory Toxicology and Pharmacology 2024-01-28

Potency determination of potential skin sensitizers in humans is essential for quantitative risk assessment and proper management. SENS-IS an vitro test based on a reconstructed human model, that was developed to predict the hazard potency sensitizers. The performance assay prediction 174 materials evaluated this work. used as benchmark determined weight evidence approach, by collectively considering all well-established data, including human, animal, chemico, vitro, silico data. Based...

10.1016/j.yrtph.2022.105128 article EN cc-by-nc-nd Regulatory Toxicology and Pharmacology 2022-01-29

Polycyclic aromatic hydrocarbons (PAHs) are environmental pollutants generated from the incomplete combustion of organic material. PAHs have been studied as genotoxicants, but some also act via non-genotoxic mechanisms in estrogen-dependent malignancies, such breast cancer. require metabolic activation to electrophilic metabolites exert their genotoxicity properties may contribute carcinogenicity. The role endometrial cancer, a cancer associated with unopposed estrogen action is unknown. We...

10.1530/joe-19-0579 article EN Journal of Endocrinology 2020-08-26

It has long been recognized that skin sensitizers either are electrophilic or can be activated to species. Several nonanimal assays for sensitization based on this premise. In the course of a project update dermal thresholds (DST), we found substantial number sensitizers, with no pro-electrophilic alerts, could simply explained in terms sensitizer acting as nucleophile. some cases, nucleophilic center is sulfur phosphorus atom, while others, it an aromatic carbon atom. For carbon-centered...

10.1021/acs.chemrestox.4c00207 article EN cc-by Chemical Research in Toxicology 2024-09-11

10.1109/isec61299.2024.10665227 article EN 2022 IEEE Integrated STEM Education Conference (ISEC) 2024-03-09

While interpretability research has shed light on some internal algorithms utilized by transformer-based LLMs, reasoning in natural language, with its deep contextuality and ambiguity, defies easy categorization. As a result, formulating clear motivating questions for circuit analysis that rely well-defined in-domain out-of-domain examples required causal interventions is challenging. Although significant work investigated circuits specific tasks, such as indirect object identification...

10.48550/arxiv.2410.21353 preprint EN arXiv (Cornell University) 2024-10-28

The U-SENS™ assay was developed to address the third key event of skin sensitization adverse outcome pathway (AOP) and is described in OECD test guideline 442E, Annex II. A dataset 68 fragrance ingredients comprised 7 non-sensitizers 61 sensitizers tested assay. potential for activate dendritic cells, measured by U-SENS™, compared determined weight evidence (WoE) from historical data. Of non-sensitizers, 4 induced CD86 cell surface marker ≥1.5-fold while 3 did not. sensitizers, 50 were...

10.1016/j.tiv.2021.105298 article EN cc-by Toxicology in Vitro 2021-12-11

Background: Digital health is a widely discussed topic in both the lay and professional press. Innovators this space continue to raise significant funds develop applications devices that will change how healthcare managed United States. Wearable mobile are used by up one third of all American consumers, presenting key opportunity give put into consumers’ hands. However, who have biggest role remains be seen. Discussion: Technology giants, biopharma firms, startups chance take advantage new...

10.2174/2213809905666180816115016 article EN Technology Transfer and Entrepreneurship 2018-08-23

In a plethora of recent work, large language models (LLMs) demonstrated impressive reasoning ability, but many proposed downstream tasks focus on performance-wise evaluation. Two fundamental questions persist: 1) how reliable is the quality reasoning, and 2) can detect unreliable reasoning? this paper, we investigate self-contradictory (Self-Contra) where model does not support predictions. To address 1), assess Self-Contra rate across four datasets delve into finer-grained categories...

10.48550/arxiv.2311.09603 preprint EN cc-by arXiv (Cornell University) 2023-01-01

Augmenting a language model (LM) with $k$-nearest neighbors ($k$NN) retrieval on its training data alone can decrease perplexity, though the underlying reasons for this remain elusive. In work, we rule out one previously posited possibility -- "softmax bottleneck." We then create new dataset to evaluate LM generalization ability in setting where contains additional information that is not causally relevant. This task challenging even GPT-3.5 Turbo. show that, both GPT-2 and Mistral 7B, $k$NN...

10.48550/arxiv.2311.09615 preprint EN other-oa arXiv (Cornell University) 2023-01-01
Coming Soon ...