Salvato in:
Dettagli Bibliografici
Autori principali: Daniali, Maryam, Karandikar, Shivaram, Zimmerman, Dabriel, Schmitt, J. Eric, Buczek, Matthew J., Jung, Benjamin, Mercedes, Laura, Seidlitz, Jakob, Troiani, Vanessa, Dorfschmidt, Lena, Kafadar, Eren, Williams, Remo, Sotardi, Susan, Vossough, Arastoo, Haag, Scott, Schabdach, Jenna M., Alexander-Bloch, Aaron
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2503.12143
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908318366171136
author Daniali, Maryam
Karandikar, Shivaram
Zimmerman, Dabriel
Schmitt, J. Eric
Buczek, Matthew J.
Jung, Benjamin
Mercedes, Laura
Seidlitz, Jakob
Troiani, Vanessa
Dorfschmidt, Lena
Kafadar, Eren
Williams, Remo
Sotardi, Susan
Vossough, Arastoo
Haag, Scott
Schabdach, Jenna M.
Alexander-Bloch, Aaron
author_facet Daniali, Maryam
Karandikar, Shivaram
Zimmerman, Dabriel
Schmitt, J. Eric
Buczek, Matthew J.
Jung, Benjamin
Mercedes, Laura
Seidlitz, Jakob
Troiani, Vanessa
Dorfschmidt, Lena
Kafadar, Eren
Williams, Remo
Sotardi, Susan
Vossough, Arastoo
Haag, Scott
Schabdach, Jenna M.
Alexander-Bloch, Aaron
contents Clinically acquired brain MRIs and radiology reports are valuable but underutilized resources due to the challenges of manual analysis and data heterogeneity. We developed fine-tuned language models (LMs) to classify brain MRI reports as normal (reports with limited pathology) or abnormal, fine-tuning BERT, BioBERT, ClinicalBERT, and RadBERT on 44,661 reports. We also explored the reasoning capabilities of a leading LM, Gemini 1.5-Pro, for normal report categorization. Automated image processing and modeling generated brain growth charts from LM-classified normal scans, comparing them to human-derived charts. Fine-tuned LMs achieved high classification performance (F1-Score >97%), with unbalanced training mitigating class imbalance. Performance was robust on out-of-distribution data, with full text outperforming summary (impression) sections. Gemini 1.5-Pro showed a promising categorization performance, especially with clinical inference. LM-derived brain growth charts were nearly identical to human-annotated charts (r = 0.99, p < 2.2e-16). Our LMs offer scalable analysis of radiology reports, enabling automated classification of brain MRIs in large datasets. One application is automated generation of brain growth charts for benchmarking quantitative image features. Further research is needed to address data heterogeneity and optimize LM reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2503_12143
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Language Models for Automated Classification of Brain MRI Reports and Growth Chart Generation
Daniali, Maryam
Karandikar, Shivaram
Zimmerman, Dabriel
Schmitt, J. Eric
Buczek, Matthew J.
Jung, Benjamin
Mercedes, Laura
Seidlitz, Jakob
Troiani, Vanessa
Dorfschmidt, Lena
Kafadar, Eren
Williams, Remo
Sotardi, Susan
Vossough, Arastoo
Haag, Scott
Schabdach, Jenna M.
Alexander-Bloch, Aaron
Image and Video Processing
Artificial Intelligence
Machine Learning
Clinically acquired brain MRIs and radiology reports are valuable but underutilized resources due to the challenges of manual analysis and data heterogeneity. We developed fine-tuned language models (LMs) to classify brain MRI reports as normal (reports with limited pathology) or abnormal, fine-tuning BERT, BioBERT, ClinicalBERT, and RadBERT on 44,661 reports. We also explored the reasoning capabilities of a leading LM, Gemini 1.5-Pro, for normal report categorization. Automated image processing and modeling generated brain growth charts from LM-classified normal scans, comparing them to human-derived charts. Fine-tuned LMs achieved high classification performance (F1-Score >97%), with unbalanced training mitigating class imbalance. Performance was robust on out-of-distribution data, with full text outperforming summary (impression) sections. Gemini 1.5-Pro showed a promising categorization performance, especially with clinical inference. LM-derived brain growth charts were nearly identical to human-annotated charts (r = 0.99, p < 2.2e-16). Our LMs offer scalable analysis of radiology reports, enabling automated classification of brain MRIs in large datasets. One application is automated generation of brain growth charts for benchmarking quantitative image features. Further research is needed to address data heterogeneity and optimize LM reasoning.
title Language Models for Automated Classification of Brain MRI Reports and Growth Chart Generation
topic Image and Video Processing
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2503.12143