AI Article Synopsis
- Accurate identification of primary central nervous system lymphoma (PCNSL) during surgery is crucial for effective neurosurgical decisions, but pathologists struggle to differentiate it from other brain lesions like glioma using frozen sections.
- Researchers developed a deep learning model named LGNet to improve this differentiation, utilizing H&E-stained images, and benchmarked its accuracy against pathologists with varying experience.
- LGNet demonstrated exceptional performance, achieving high accuracy metrics (AUROCs) in distinguishing PCNSL from glioma and non-PCNSL lesions, significantly surpassing many pathologists, highlighting its potential as a vital tool in surgical diagnoses.
Article Abstract
Accurate intraoperative differentiation of primary central nervous system lymphoma (PCNSL) remains pivotal in guiding neurosurgical decisions. However, distinguishing PCNSL from other lesions, notably glioma, through frozen sections challenges pathologists. Here we sought to develop and validate a deep learning model capable of precisely distinguishing PCNSL from non-PCNSL lesions, especially glioma, using hematoxylin and eosin (H&E)-stained frozen whole-slide images. Also, we compared its performance against pathologists of varying expertise. Additionally, a human-machine fusion approach integrated both model and pathologic diagnostics. In external cohorts, LGNet achieved AUROCs of 0.965 and 0.972 in distinguishing PCNSL from glioma and AUROCs of 0.981 and 0.993 in differentiating PCNSL from non-PCNSL lesions. Outperforming several pathologists, LGNet significantly improved diagnostic performance, further augmented to some extent by fusion approach. LGNet's proficiency in frozen section analysis and its synergy with pathologists indicate its valuable role in intraoperative diagnosis, particularly in discriminating PCNSL from glioma, alongside other lesions.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11069536 | PMC |
| http://dx.doi.org/10.1038/s41467-024-48171-x | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
68Ga-Pentixafor PET in Combination With MRI Improves the Differential Diagnosis of Glioblastoma and Primary Central Nervous System Lymphoma.
Clin Nucl Med
January 2025
Department of Hematology, the First Affiliated Hospital, Fujian Medical University, Fuzhou, China.
Purposes: This study aims to investigate the diagnostic performance of combining 68Ga-pentixafor PET with MRI to differentiate primary central nervous system lymphoma (PCNSL) from glioblastoma (GBM), particularly focusing on atypical lymphoma identification.
Patients And Methods: Seventy-one PCNSL and 53 GBM patients who underwent both 68Ga-pentixafor PET/CT and MRI were retrospectively included. We evaluated the quantitative imaging parameters and MRI features of positive lesions, identifying atypical PCNSL by hemorrhage, necrosis, or heterogeneous enhancement.
MRI-based radiomics virtual biopsy for BCL6 in primary central nervous system lymphoma.
Clin Radiol
January 2025
Department of Radiology, Huashan Hospital, Fudan University, Shanghai, China. Electronic address:
Article Synopsis
- The study aimed to create a machine learning model that uses radiomic signatures to predict BCL-6 rearrangement in primary central nervous system lymphoma (PCNSL).
- Researchers analyzed imaging data from 102 PCNSL patients, extracting features from MRI scans to build and evaluate various machine learning models, with logistic regression showing the best performance.
- The findings indicate that radiomics signatures, particularly from contrast-enhanced T1-weighted and fluid-attenuated inversion recovery imaging, are effective in distinguishing BCL-6 rearrangement, providing a valuable tool for personalized patient predictions.
Distinguishing clinical and imaging characteristics of primary central nervous system lymphoma from high-grade glioma and metastatic brain tumors.
Clin Transl Oncol
October 2024
Department of Hematology, the Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, China.
Objective: The purpose of this retrospective analysis was to evaluate the clinical presentations, radiological characteristics, patient outcomes, and therapeutic approaches among individuals diagnosed with primary central nervous system lymphoma (PCNSL), high-grade glioma (HGG), and metastatic brain tumors (METS).
Methods: We assembled a cohort of brain tumor patients from two medical centers, with two oncologists independently reviewing their clinical profiles. A retrospective examination of 87 PCNSL, 87 HGG, and 71 METS cases was performed to assess the aforementioned parameters.
Primary Intraocular Lymphoma: Rad-Path and Ophthalmologic Correlation.
AJNR Am J Neuroradiol
September 2024
(1) University of Rochester, School of Medicine and Dentistry, Rochester NY 14620 (D.S), (2) Eastern Virginia Medical School, Norfolk, VA 23507 (J.P), (3) University of Connecticut, School of Medicine, Farmington, CT 06032 (ER), (4) Department of Radiology, Mayo Clinic, Jacksonville, 4500 San Pablo Road, Jacksonville, FL 55902, USA (P.V, A.A, N.S), (5) Department of Radiology, Mayo Clinic, 200 1st Street SW, Rochester, MN 55902, USA (G.B), (6) Department of Pathology, Mayo Clinic, Jacksonville, 4500 San Pablo Road, Jacksonville, FL 55902, USA (S.W).
Primary intraocular lymphoma (PIOL) is a rare form of primary central nervous system lymphoma that poses diagnostic challenges due to its nonspecific clinical features and complex imaging characteristics. This paper presents a focus case and two companion cases, highlighting the complexities in identifying and treating PIOL. In the focus case, A 66-year-old male experienced gradual painless vision loss with choroidal thickening on funduscopic exam and subsequent follow-up MRI.
View Article and Find Full Text PDFMulticenter investigation of preoperative distinction between primary central nervous system lymphomas and glioblastomas through interpretable artificial intelligence models.
Neuroradiology
November 2024
Laboratory for Medical Imaging Informatics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai, 200083, China.
Objective: Research into the effectiveness and applicability of deep learning, radiomics, and their integrated models based on Magnetic Resonance Imaging (MRI) for preoperative differentiation between Primary Central Nervous System Lymphoma (PCNSL) and Glioblastoma (GBM), along with an exploration of the interpretability of these models.
Materials And Methods: A retrospective analysis was performed on MRI images and clinical data from 261 patients across two medical centers. The data were split into a training set (n = 153, medical center 1) and an external test set (n = 108, medical center 2).
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!