Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 12 min ago
Peer-reviewedCC BY 4.0JACC: Advances

INSIGHT Model

  • Akshay Arora1
  • Muhammad Shahzeb Khan1,2,3
  • Cynthia Sunderman1
  • Nicole Baldwin1
  • Monica Bennett1
  • Vee Anne Bomar1
  • Lisa Cross1
  • Briget da Graca1
  • Jason Ettlinger1
  • Robert L. Gottlieb1,3,4,5,6,7
  • Kendall Hammonds1
  • Michael Mack1
6 more authors
  • Ahmad Masri8
  • Mia Papas9
  • Elisa L. Priest1
  • Courtney Shaver1
  • John Venditto9
  • Javed Butler1,10
10 affiliations
  1. 1Baylor Scott & White Research Institute, Dallas, Texas, USA
  2. 2Department of Medicine, Baylor College of Medicine, Temple, Texas, USA
  3. 3Baylor Scott & White The Heart Hospital Plano, Plano, Texas, USA
  4. 4Baylor Scott & White The Heart Hospital, Dallas, Texas, USA
  5. 5Baylor University Medical Center, Dallas, Texas, USA
  6. 6Texas A&M Health Science Center, Dallas, Texas, USA
  7. 7Burnett School of Medicine at TCU, Fort Worth, Texas, USA
  8. 8OHSU-PSU School of Public Health, Oregon Health and Science University, Portland, Oregon, USA
  9. 9US Medical Affairs, BioPharmaceutical Medical, AstraZeneca, Wilmington, Delaware, USA
  10. 10University of Mississippi Medical Center, Jackson, Mississippi, USA
Read full textPDFDOICite

Abstract

Background: Transthyretin amyloidosis (ATTR) diagnoses are frequently delayed or missed, attenuating treatment benefit.

Objectives: The authors developed an explainable artificial intelligence screening tool to identify high-risk patients who should be considered for diagnostic testing for ATTR-cardiomyopathy (ATTR-CM) from an unselected heart failure (HF) population.

Methods: Electronic health record data (April 2020–March 2023) were used to identify patients with suspected ATTR-CM in an HF cohort. Natural language processing with Named Entity Recognition processed unstructured clinical, echocardiographic, and electrocardiographic data. Clinical ModernBERT predicted suspected ATTR-CM. Local interpretable model-agnostic explanations were provided. The authors examined sensitivity, specificity, positive predictive value, negative predictive value, F1 score, area under the receiver-operating characteristic curve, and key model predictors.

Results: This study included 100,290 patients with HF (mean age: 71.6 ± 14.3 years; 47.3% female; 76.2% White; 18.2% Black; 89.7% non-Hispanic). The Clinical ModernBERT model, INSIGHT-CM, identified suspected ATTR-CM (n = 1,108; prevalence 1.2%) with sensitivity 77.5% (95% CI: 75.9%-82.8%), specificity 94.7% (95% CI: 94.4%-95.0%), positive predictive value 19.9% (95% CI: 17.6%-22.36%), negative predictive value 99.6% (95% CI: 99.5%-99.7%), and area under the receiver-operating characteristic curve 92.4% (95% CI: 90.5%-94.3%).

Conclusions: In an unselected HF population with 1% prevalence of suspected ATTR-CM, INSIGHT-CM identified patients among whom 1 in 5 fit the pattern for suspected ATTR-CM, offering meaningful opportunities to improve timely diagnosis and treatment.

Cite

BibTeXRIS
BibTeX
@article{arora2026insight,
  title = {{INSIGHT Model}},
  author = {Akshay Arora and Muhammad Shahzeb Khan and Cynthia Sunderman and Nicole Baldwin and Monica Bennett and Vee Anne Bomar and Lisa Cross and Briget da Graca and Jason Ettlinger and Robert L. Gottlieb and Kendall Hammonds and Michael Mack and Ahmad Masri and Mia Papas and Elisa L. Priest and Courtney Shaver and John Venditto and Javed Butler},
  journal = {JACC: Advances},
  year = {2026},
  month = aug,
  doi = {10.1016/j.jacadv.2026.103122},
  url = {https://doi.org/10.1016/j.jacadv.2026.103122},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - INSIGHT Model
AU  - Akshay Arora
AU  - Muhammad Shahzeb Khan
AU  - Cynthia Sunderman
AU  - Nicole Baldwin
AU  - Monica Bennett
AU  - Vee Anne Bomar
AU  - Lisa Cross
AU  - Briget da Graca
AU  - Jason Ettlinger
AU  - Robert L. Gottlieb
AU  - Kendall Hammonds
AU  - Michael Mack
AU  - Ahmad Masri
AU  - Mia Papas
AU  - Elisa L. Priest
AU  - Courtney Shaver
AU  - John Venditto
AU  - Javed Butler
PY  - 2026
DA  - 2026/08/26
JO  - JACC: Advances
DO  - 10.1016/j.jacadv.2026.103122
UR  - https://doi.org/10.1016/j.jacadv.2026.103122
AB  - Background: Transthyretin amyloidosis (ATTR) diagnoses are frequently delayed or missed, attenuating treatment benefit.  Objectives: The authors developed an explainable artificial intelligence screening tool to identify high-risk patients who should be considered for diagnostic testing for ATTR-cardiomyopathy (ATTR-CM) from an unselected heart failure (HF) population.  Methods: Electronic health record data (April 2020–March 2023) were used to identify patients with suspected ATTR-CM in an HF cohort. Natural language processing with Named Entity Recognition processed unstructured clinical, echocardiographic, and electrocardiographic data. Clinical ModernBERT predicted suspected ATTR-CM. Local interpretable model-agnostic explanations were provided. The authors examined sensitivity, specificity, positive predictive value, negative predictive value, F1 score, area under the receiver-operating characteristic curve, and key model predictors.  Results: This study included 100,290 patients with HF (mean age: 71.6 ± 14.3 years; 47.3% female; 76.2% White; 18.2% Black; 89.7% non-Hispanic). The Clinical ModernBERT model, INSIGHT-CM, identified suspected ATTR-CM (n = 1,108; prevalence 1.2%) with sensitivity 77.5% (95% CI: 75.9%-82.8%), specificity 94.7% (95% CI: 94.4%-95.0%), positive predictive value 19.9% (95% CI: 17.6%-22.36%), negative predictive value 99.6% (95% CI: 99.5%-99.7%), and area under the receiver-operating characteristic curve 92.4% (95% CI: 90.5%-94.3%).  Conclusions: In an unselected HF population with 1% prevalence of suspected ATTR-CM, INSIGHT-CM identified patients among whom 1 in 5 fit the pattern for suspected ATTR-CM, offering meaningful opportunities to improve timely diagnosis and treatment.
SN  - 2772-963X
ER  -