Alexander T. Pearson

Alexander T. Pearson

Auto-generated profile, based on publicly available information

Alexander T. Pearson

Department of Medicine, Section of Hematology/Oncology, University of Chicago Medical Center, 5841 South Maryland Avenue, MC 2115, Chicago, IL 60637, USA; University of Chicago Comprehensive Cancer Center, Chicago, IL, USA; The University of Chicago Medicine & Biological Sciences, 5841 South Maryland Avenue, Chicago, IL, USA

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 July 2020 in Nature Cancer
Verified
Journal Article
Closed access
PDF file
JKJakob Nikolas Kather
Jakob Nikolas Kather
LHLara R. Heij
Lara R. Heij
HGHeike Irmgard Grabsch
Heike Irmgard Grabsch
CLChiara Loeffler
Chiara Loeffler
AEAmelie Echle
Amelie Echle
TLTom Lüdde
Tom Lüdde

20 July 2021 in Nature Communications
Verified
Journal Article
Open access
PDF file
FHFrederick M. Howard
Frederick M. Howard
JDJames Dolezal
James Dolezal
SKSara Kochanny
Sara Kochanny
JSJefree Schulte
Jefree Schulte
HCHeather Chen
Heather Chen
APAlexander T. Pearson
Alexander T. Pearson
The Cancer Genome Atlas (TCGA) is one of the largest biorepositories of digital histology. Deep learning (DL) models have been trained on TCGA to predict numerous features directly from histology, including survival, gene expression patterns, and driver mutations. However, we demonstrate that these features vary substantially across tissue submitting sites in TCGA for over 3,000 patients with six cancer subtypes. Additionally, we show that histologic image differences between submitting sites can easily be identified with DL. Site detection remains possible despite commonly used color normalization and augmentation methods, and we quantify the image characteristics constituting this site-specific digital histology signature. We demonstrate that these site-specific signatures lead to biased accuracy for prediction of features including survival, genomic mutations, and tumor stage. Furthermore, ethnicity can also be inferred from site-specific signatures, which must be accounted for to ensure equitable application of DL. These site-specific signatures can lead to overoptimistic estimates of model performance, and we propose a quadratic programming method that abrogates this bias by ensuring models are not trained and validated on samples from the same site.

27 December 2022 in bioRxiv
Verified
Preprint
Open access
PDF file
CGCatherine A. Gao
Catherine A. Gao
FHFrederick M. Howard
Frederick M. Howard
NMNikolay S. Markov
Nikolay S. Markov
EDEmma C. Dyer
Emma C. Dyer
SRSiddhi Ramesh
Siddhi Ramesh
APAlexander T. Pearson
Alexander T. Pearson
Background Large language models such as ChatGPT can produce increasingly realistic text, with unknown information on the accuracy and integrity of using these models in scientific writing. Methods We gathered ten research abstracts from five high impact factor medical journals (n=50) and asked ChatGPT to generate research abstracts based on their titles and journals. We evaluated the abstracts using an artificial intelligence (AI) output detector, plagiarism detector, and had blinded human reviewers try to distinguish whether abstracts were original or generated. Results All ChatGPT-generated abstracts were written clearly but only 8% correctly followed the specific journal’s formatting requirements. Most generated abstracts were detected using the AI output detector, with scores (higher meaning more likely to be generated) of median [interquartile range] of 99.98% [12.73, 99.98] compared with very low probability of AI-generated output in the original abstracts of 0.02% [0.02, 0.09]. The AUROC of the AI output detector was 0.94. Generated abstracts scored very high on originality using the plagiarism detector (100% [100, 100] originality). Generated abstracts had a similar patient cohort size as original abstracts, though the exact numbers were fabricated. When given a mixture of original and general abstracts, blinded human reviewers correctly identified 68% of generated abstracts as being generated by ChatGPT, but incorrectly identified 14% of original abstracts as being generated. Reviewers indicated that it was surprisingly difficult to differentiate between the two, but that the generated abstracts were vaguer and had a formulaic feel to the writing. Conclusion ChatGPT writes believable scientific abstracts, though with completely generated data. These are original without any plagiarism detected but are often identifiable using an AI output detector and skeptical human reviewers. Abstract evaluation for journals and medical conferences must adapt policy and practice to maintain rigorous scientific standards; we suggest inclusion of AI output detectors in the editorial process and clear disclosure if these technologies are used. The boundaries of ethical and acceptable use of large language models to help scientific writing remain to be determined.

2 November 2022 in Nature Communications
Verified
Journal Article
Open access
PDF file
JDJames M. Dolezal
James M. Dolezal
ASAndrew Srisuwananukorn
Andrew Srisuwananukorn
DKDmitry Karpeyev
Dmitry Karpeyev
SRSiddhi Ramesh
Siddhi Ramesh
SKSara Kochanny
Sara Kochanny
APAlexander T. Pearson
Alexander T. Pearson
A model’s ability to express its own predictive uncertainty is an essential attribute for maintaining clinical user confidence as computational biomarkers are deployed into real-world medical settings. In the domain of cancer digital histopathology, we describe a clinically-oriented approach to uncertainty quantification for whole-slide images, estimating uncertainty using dropout and calculating thresholds on training data to establish cutoffs for low- and high-confidence predictions. We train models to identify lung adenocarcinoma vs. squamous cell carcinoma and show that high-confidence predictions outperform predictions without uncertainty, in both cross-validation and testing on two large external datasets spanning multiple institutions. Our testing strategy closely approximates real-world application, with predictions generated on unsupervised, unannotated slides using predetermined thresholds. Furthermore, we show that uncertainty thresholding remains reliable in the setting of domain shift, with accurate high-confidence predictions of adenocarcinoma vs. squamous cell carcinoma for out-of-distribution, non-lung cancer cohorts.

4 May 2010 in Neurology
Verified
Journal Article
Closed access
ELE.L. Logigian
E.L. Logigian
WMW.B. Martens
W.B. Martens
RMR.T. Moxley
R.T. Moxley
MMM.P. McDermott
M.P. McDermott
NDN. Dilek
N. Dilek
CTC.A. Thornton
C.A. Thornton
To determine if mexiletine is safe and effective in reducing myotonia in myotonic dystrophy type 1 (DM1). Myotonia is an early, prominent symptom in DM1 and contributes to decreased dexterity, gait instability, difficulty with speech/swallowing, and muscle pain. A few preliminary trials have suggested that the antiarrhythmic drug mexiletine is useful, symptomatic treatment for nondystrophic myotonic disorders and DM1. We performed 2 randomized, double-blind, placebo-controlled crossover trials, each involving 20 ambulatory DM1 participants with grip or percussion myotonia on examination. The initial trial compared 150 mg of mexiletine 3 times daily to placebo, and the second trial compared 200 mg of mexiletine 3 times daily to placebo. Treatment periods were 7 weeks in duration separated by a 4- to 8-week washout period. The primary measure of myotonia was time for isometric grip force to relax from 90% to 5% of peak force after a 3-second maximum grip contraction. EKG measurements and adverse events were monitored in both trials. There was a significant reduction in grip relaxation time with both 150 and 200 mg dosages of mexiletine. Treatment with mexiletine at either dosage was not associated with any serious adverse events, or with prolongation of the PR or QTc intervals or of QRS duration. Mild adverse events were observed with both placebo and mexiletine treatment. Mexiletine at dosages of 150 and 200 mg 3 times daily is effective, safe, and well-tolerated over 7 weeks as an antimyotonia treatment in DM1. This study provides Class I evidence that mexiletine at dosages of 150 and 200 mg 3 times daily over 7 weeks is well-tolerated and effective in reducing handgrip relaxation time in DM1.

2 November 2020 in JAMA Network Open
Verified
Journal Article
Closed access
PDF file
FHFrederick Matthew Howard
Frederick Matthew Howard
SKSara Kochanny
Sara Kochanny
MKMatthew Koshy
Matthew Koshy
MSMichael Spiotto
Michael Spiotto
APAlexander T. Pearson
Alexander T. Pearson
The adjuvant treatment of resected head and neck squamous cell carcinoma (HNSCC) is guided by the European Organisation for Research and Treatment of Cancer (EORTC) 22931 and Radiation Therapy Oncology Group (RTOG) 95-01 randomized clinical trials.1,2 Both trials randomized patients with at least 1 adverse prognostic factor after definitive surgery to adjuvant radiotherapy (RT) or cisplatin-based chemoradiotherapy (CRT), and both demonstrated a progression-free survival (PFS) benefit with CRT. The specific adverse risk features qualifying patients for each trial were variable, but an exploratory combined analysis3 suggested that extracapsular extension (ECE) and positive surgical margins were the most significant prognostic factors, and patients with 1 of these 2 features derived benefit from CRT.