Clinical and Translational Scientist Seminar Series

LEARN'd Conference

Dashboard and Mobile Health Application

Ohio Equity Initiative

Available per PI request
A set of dashboards hosted by the Ohio Department of Medicaid to display health outcomes on infant mortality among high-risk pregnant women across the state of Ohio.
Reference: PMID: 34974826
Contact: Naleef Fareed
Achieve Study

Available per PI request.
A clinical dashboard embedded in EPIC for the maternal fetal medicine and endocrinology care team to use for monitoring health and social needs progress among pregnant individuals living with T2D.
Reference: PMID: 37164461
Contact: Naleef Fareed
HEALing Communities Study

Available per PI request.
A NIH funded dashboard that provided user-based access to opioid use health outcomes across the state of Ohio to community and study end-users involved in the HEALing Communities Study. The dashboard provides information to base decisions on selection of evidence-based practices within a community.
Reference: PMID: 33070058
Contact: Naleef Fareed
Integrated Behavioral Health

A dashboard hosted by the state of Ohio displaying opioid and substance abuse related health outcomes across all 88 counties in the state.
Contact: Naleef Fareed
Ohio Opportunity Index

A dashboard displaying area deprivation information across all census tracts in the state of Ohio; used by the state department of Medicaid for resource allocation
Reference: PMID: 32757202
Contact: Naleef Fareed
Ohio Children's Opportunity Index

A dashboard displaying area deprivation information across all census tracts in the state of Ohio; used by the state department of Medicaid for resource allocation focused on children's health outcomes.
Reference: PMID: 35081656
Contact: Naleef Fareed
Github Repository Ohio Opportunity Index

A repository containing R scripts and teaching material to develop area deprivation indices and build dashbaords in Tableau.
Reference: PMID: 35942261
Contact: Naleef Fareed
Github Repository HEALing Communities Study

A repository containing software scripts (R, Drupal) and teaching material to develop substance use measures and build dashboards in various softwares.
Contact: Naleef Fareed
Achieve Study

Available per PI request.
A mobile health application used by pregnant individuals living with T2D to report continous glucose values (via a Dexcom device) and report momentary ecological assessments on social needs, diabetes distress and related health outcomes.
Reference: PMID: 36644662
Contact: Naleef Fareed

Image Data Analytics

ITK-SNAP

ITK-SNAP is an open-source software application for medical image segmentation. It features manual, semi-automatic and automatic tools for delineating anatomical structures in 3D imaging data.
Reference: https://doi.org/10.1016/j.neuroimage.2006.01.015
Contact: Yuan Xue
ANTs

ANTs is a suite of image processing tools for medical image analysis, including image registration, segmentation and statistical analysis. It is widely used for brain imaging studies and other biomedical applications.
Reference: https://doi.org/10.1038/s41598-021-87564-6
Contact: Yuan Xue
3D Slicer

3D Slicer is a free, open-source software platform for the analysis and visualization of medical images. It supports a variety of image processing tasks, including segmentation, registration and 3D reconstruction.
Reference: PMID: 22770690
Contact: Yuan Xue
Segment Anything Model 2 (SAM 2)

SAM 2 is a general-purpose, promptable segmentation model that can be used to segment objects in images and videos. Developed by Meta AI, it is designed to handle a wide variety of image types and segmentation tasks with minimal manual annotation.
Reference: ArXiv: 2408.00714
Contact: Yuan Xue
QuPath

QuPath is an open-source software for digital pathology image analysis. It provides tools for the visualization, annotation, and analysis of whole slide images, with a focus on machine learning-based methods for tissue and cell classification.
Reference: https://doi.org/10.1038/s41598-017-17204-5
Contact: Yuan Xue
ImageJ

ImageJ is a public domain, Java-based image processing program developed at the National Institutes of Health (NIH). It is widely used for scientific image analysis and supports a wide range of plugins for various types of image processing tasks.
Referenece: https://doi.org/10.1186/s12859-017-1934-z
Contact: Yuan Xue
MONAI (Medical Open Network for AI)

MONAI is an open-source framework designed for deep learning in healthcare imaging. It provides domain-optimized foundational capabilities for developing AI models, including pre-built pipelines for segmentation, classification and other image analysis tasks.
Reference: ArXiv: 2211.02701
Contact: Yuan Xue
nnU-Net

nnU-Net is a self-adapting framework for biomedical image segmentation. It automatically configures itself to a given dataset, optimizing the architecture, preprocessing and training pipelines. It has been highly successful in various biomedical image segmentation challenges.
Reference: https://doi.org/10.1038/s41592-020-01008-z
Contact: Yuan Xue

All of Us Researcher Workbench Workshop at The Ohio State University

Cloud Infrastructure Development

NIH STRIDES

An NIH program to allow researchers use their NIH funding directly to pay for cloud platform with a pricing discount.
Contact: Qin Ma
Azure

Azure is the cloud computing platform developed by Microsoft. Specially, it can support with the NIH STRIDES program at The Ohio State University Wexner Medical Center.
Contact: Qin Ma
Amazon Web Services

Amazon Web Services (AWS) is the cloud computing platform developed by Amazon. Specially, it can provide PHI data analysis and storage at Ohio State.
Contact: Qin Ma
Terra

Terra is a cloud-native platform for biomedical researchers to access data, run analysis tools and collaborate.
Contact: Qin Ma
Brainlife.io

A decentralized and open-source cloud platform to support neuroscience research.
Reference: PMID: 38605111
Contact: Qin Ma
DeepCell Kiosk

A cloud-native software that dynamically scales deep learning workflows to accommodate large imaging datasets. Provided practical ways to deploy cloud deep learning infrastructure
Reference: PMID: 33398191
Contact: Qin Ma
ssREAD

A single-cell and spatial RNA-seq database for Alzheimer’s disease (ssREAD) built on cloud. It offers a broader spectrum of AD-related datasets, an optimized analytical pipeline and improved usability over cloud.
Reference: PMID: 38844475
Contact: Qin Ma

Electronic Health Record (EHR) Predictive Analytical Models

DeepIPW

Treatment Effect Estimation based on deep learning framework.
Reference: PMID: 35603127
Contact: Ping Zhang
CURE

Treatment Effect Estimation based on deep learning framework.
Reference: PMID: 39005483
Contact: Ping Zhang
KG-TREAT

Treatment Effect Estimation based on deep learning framework.
Reference: PMID: 38651015
Contact: Ping Zhang
BiLSTM

Predictive models based on deep learning.
Reference: PMID: 33659912
Contact: Ping Zhang

Pharmacology Knowledge Base

MPRINT-KP Silver

The Maternal and Pediatric Precision in Therapeutics (MPRINT) knowledge base is a crucial maternal and pediatric research informatics infrastructure. The silver version of the MPRINT knowledge base aims to assist researchers in discovering published maternal and pediatric research easily and quickly. The deep-learning model was trained using a comprehensive maternal and pediatric study corpus. Subsequently, PubMed abstracts were thoroughly screened using the established deep learning model to predict the publication study types and relevant populations. Users can retrieve studies based on the drugs, diseases, specific populations, and study types.
Reference: PMID: 36625779, https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.2835
Contact: Lang Li
MPRINT-KP Gold

The Maternal and Pediatric Precision in Therapeutics (MPRINT) knowledge base is a crucial maternal and pediatric research informatics infrastructure. The gold version of the MPRINT knowledge base aims to provide expert-curated knowledge from published maternal and pediatric studies. It primarily provides drug pharmacokinetic parameters and epidemiological evidence. Each piece of knowledge is well-annotated with relevant information, including demographic information, therapeutics, and other risk factors.
Reference: PMID: 36625779, https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.2835
Contact: Lang Li
DrugCombo

DrugCombo is a comprehensive knowledge base that integrates drug toxicity and pharmacokinetic data for single drugs and drug combinations from the United States Food and Drug Administration Adverse Event Report System (FAERS), drug labeling from the National Library of Medicine’s DailyMed website, published study results in PubMed, and the DrugBank database. DrugCombo is the first database to contain crucial data, such as the maximum tolerable dose (MTD), dose-limiting toxicity (DLT), and dose range, for Phase I clinical trial design. It also provides pharmacokinetic evidence of drug interactions between cancer drugs and other relevant knowledge. DrugCombo currently integrates 8577 drug/target interactions and 8797 drug interactions from DrugBank, 3995 severe adverse drug events (ADEs), 95,535 common ADEs from drug labels, 1,816,030 ADEs from FAERS, and MTD and DLTs from 2592 Phase I trials.
Reference: https://doi.org/10.1158/1538-7445.AM2023-6573 , https://doi.org/10.1200/JCO.2023.41.16_suppl.1571
Contact: Lang Li
SLKB

Synthetic lethality knowledge base (SLKB) is dedicated to curating CRISPR double knockout experiments (CDKO) aiming to identify synthetic lethal (SL) interactions between two genes. SL identification is highly context dependent, differing across pathways, gene targets, cell lines, and CDKO libraries. SLKB analysis pipeline, additionally distributed as a python package, allows SL score calculation for CDKO studies. Via SLKB, users can analyze their own CDKO data and browse their results.
Reference: PMID: 37889037
Contact: Lang Li
CTKG

CTKG is a knowledge graph integrating a wide variety of information from clinical trials (e.g., interventions, drug side effects, disease symptoms). Together with computational tools to learn from knowledge graphs, CTKG faciliates knowledge discovery from clinical trials, and many relevant tasks such as information retrieval, drug repurposing, etc.
Reference: PMID: 35304504
Contact: Xia Ning, Access Ning Lab
DRKG

DRKG is a very large and comprehensive knowledge graph integrating a wide variety of entities, including drugs, diseases, symptoms, genes, proteins, etc. Together with computational tools to learn from knowledge graphs, DRKG facilitates drug repurposing, among other tasks such as information retrieval, etc.
Reference: https://github.com/gnn4dr/DRKG/blob/master/DRKG%20Drug%20Repurposing%20Knowledge%20Graph.pdf
Contact: Xia Ning

Natural Language Processing (NLP)

BioBERT

BioBERT is Bidirectional Encoder Representations from Transformers for Biomedical Text Mining. It is a domain-specific language representation model pre-trained on large-scale biomedical corpora. BioBERT significantly outperforms many text mining methods on the following three representative biomedical text mining tasks: biomedical named entity recognition, biomedical relation extraction and biomedical question answering.
Reference: PMID: 31501885
Contact: Lang Li
BERT

BERT is a transformer-based model that employs bidirectional training of Transformer encoders to achieve state-of-the-art performance in numerous natural language processing tasks at the time of its introduction. The development of BERT laid the groundwork for current generations of large language models, serving as a proof of concept for bidirectional transformer architectures. BERT is built upon the attention mechanism but deviates from the original encoder-decoder structure by focusing solely on the encoder. Its architecture consists of multiple layers of bidirectional self-attention and is pre-trained on a large corpus of text, including the entire Wikipedia and BookCorpus. BERT's key innovation is its ability to capture context from both directions, allowing for a deeper understanding of language semantics.
Reference: https://arxiv.org/abs/1810.04805
Contact: Lang Li
PubBERT

PubBERT is a variant of BERT specifically fine-tuned on a corpus of biomedical literature, such as PubMed articles, to enhance its performance in biomedical text processing tasks. This adaptation allows PubBERT to better understand and generate language pertinent to the biomedical field, improving upon the general BERT model for domain-specific applications.
Contact: Lang Li
Llama 2

Llama 2 is a series of large language models by Meta AI, designed for various NLP tasks, available in three sizes: 7 billion, 13 billion and 70 billion parameters.
Reference: https://arxiv.org/abs/2207.06947
Contact: Ping Zhang
Mistral

Mistral is an advanced language model designed for various natural language processing tasks, available in two sizes: 7 billion and 8x7 billion parameters.
Reference: https://arxiv.org/abs/2307.02022
Contact: Steve Rust
Mixtral

Mixtral combines multiple 7 billion parameter models to enhance performance across various NLP tasks.
Reference: https://arxiv.org/abs/2307.02022
Contact: Steve Rust
Phi 2

Phi 2 is a powerful language model designed for natural language understanding and generation.
Reference: https://arxiv.org/abs/2306.11692
Contact: Steve Rust
Flan-T5

Flan-T5 is a fine-tuned language model from Google, optimized for various NLP tasks, available in sizes Lg 780M and XXL 11B.
Reference: https://arxiv.org/abs/2210.11416
Contact: Steve Rust
Falcon

Falcon is a high-performance language model optimized for speed and efficiency, available in sizes 7 billion and 40 billion parameters.
Reference: https://arxiv.org/abs/2208.13269
Contact: Steve Rust
GatorTron

GatorTron is a biomedical language model developed by the University of Florida, available in sizes S 345M, Medium 3.9B, and Large 9B.
Reference: https://arxiv.org/abs/2207.07177
Contact: Steve Rust
RoBERTa

RoBERTa (Robustly optimized BERT approach) is a transformer-based model that improves on BERT by training with larger mini-batches and more data.
Reference: https://arxiv.org/abs/1907.11692
Contact: Steve Rust
Phi 3

Phi 3 is the latest iteration of the Phi series, designed for advanced natural language processing and understanding.
Reference: https://arxiv.org/abs/2308.11239
Contact: Steve Rust
MPT

MPT is a versatile transformer model available in sizes 7 billion and 30 billion parameters, designed for a wide range of NLP tasks.
Reference: https://arxiv.org/abs/2307.11590
Contact: Steve Rust
OPT

OPT (Open Pre-trained Transformer) is a series of transformer models designed for various NLP tasks, available in multiple sizes.
Reference: https://arxiv.org/abs/2205.01068
Contact: Steve Rust

Clinical and Translational Science Pilot Program Ideation Session 2

Subscribe to