Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models

<b>Background:</b> Malignant pleural effusions (MPEs) are common in advanced lung cancer patients. Cytological examination of pleural fluid is essential for identifying cell types but presents diagnostic challenges, particularly when reactive mesothelial cells mimic neoplastic cells. AI-...

Full description

Saved in:
Bibliographic Details
Main Authors: Enrico Giarnieri, Elisabetta Carico, Stefania Scarpino, Alberto Ricci, Pierdonato Bruno, Simone Scardapane, Daniele Giansanti
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/10/1240
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849711825065607168
author Enrico Giarnieri
Elisabetta Carico
Stefania Scarpino
Alberto Ricci
Pierdonato Bruno
Simone Scardapane
Daniele Giansanti
author_facet Enrico Giarnieri
Elisabetta Carico
Stefania Scarpino
Alberto Ricci
Pierdonato Bruno
Simone Scardapane
Daniele Giansanti
author_sort Enrico Giarnieri
collection DOAJ
description <b>Background:</b> Malignant pleural effusions (MPEs) are common in advanced lung cancer patients. Cytological examination of pleural fluid is essential for identifying cell types but presents diagnostic challenges, particularly when reactive mesothelial cells mimic neoplastic cells. AI-powered diagnostic systems have emerged as valuable tools in digital cytopathology. This study explores the applicability of machine-learning (ML) models and highlights the importance of accessible tools for clinicians, enabling them to develop AI solutions and make advanced diagnostic tools available even in resource-limited settings. The focus is on differentiating normal/reactive cells from neoplastic cells in pleural effusions linked to lung adenocarcinoma. <b>Methods:</b> A dataset from the Cytopathology Unit at the Sant’Andrea University Hospital comprising 969 raw images, annotated with 3130 single mesothelial cells and 3260 adenocarcinoma cells, was categorized into two classes based on morphological features. Object-detection models were developed using YOLOv8 and the latest YOLOv11 instance segmentation models. <b>Results:</b> The models achieved an Intersection over Union (IoU) score of 0.72, demonstrating robust performance in class prediction for both categories, with YOLOv11 showing performance improvements over YOLOv8 in different metrics. <b>Conclusions:</b> The application of machine learning in cytopathology offers clinicians valuable support in differential diagnosis while also expanding their ability to engage with AI tools and methodologies. The diagnosis of MPEs is marked by substantial morphological and technical variability, underscoring the need for high-quality datasets and advanced deep-learning models. These technologies have the potential to enhance data interpretation and support more effective clinical treatment strategies in the era of precision medicine.
format Article
id doaj-art-5485e4043b9e4e4b8c4cf93b0f3c9111
institution DOAJ
issn 2075-4418
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-5485e4043b9e4e4b8c4cf93b0f3c91112025-08-20T03:14:31ZengMDPI AGDiagnostics2075-44182025-05-011510124010.3390/diagnostics15101240Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly ModelsEnrico Giarnieri0Elisabetta Carico1Stefania Scarpino2Alberto Ricci3Pierdonato Bruno4Simone Scardapane5Daniele Giansanti6Cytopathology Unit, Department of Clinical and Molecular Medicine, Sant’Andrea University Hospital, Sapienza University of Rome, Via di Grottarossa 1035, 00189 Rome, ItalyCytopathology Unit, Department of Clinical and Molecular Medicine, Sant’Andrea University Hospital, Sapienza University of Rome, Via di Grottarossa 1035, 00189 Rome, ItalyMorphologic and Molecular Pathology Unit, Department of Clinical and Molecular Medicine, Sant’ Andrea University Hospital, Sapienza University of Rome, Via di Grottarossa 1035, 00189 Rome, ItalyRespiratory Disease Unit, Sant’Andrea University Hospital, Sapienza University of Rome, Via di Grottarossa 1035, 00189 Rome, ItalyRespiratory Disease Unit, Sant’Andrea University Hospital, Sapienza University of Rome, Via di Grottarossa 1035, 00189 Rome, ItalyDepartment of Information Engineering, Electronics and Telecommunications, Sapienza University of Rome, Via Eudossiana 18, 00196 Rome, ItalyCentre Tisp, ISS, 00161 Rome, Italy<b>Background:</b> Malignant pleural effusions (MPEs) are common in advanced lung cancer patients. Cytological examination of pleural fluid is essential for identifying cell types but presents diagnostic challenges, particularly when reactive mesothelial cells mimic neoplastic cells. AI-powered diagnostic systems have emerged as valuable tools in digital cytopathology. This study explores the applicability of machine-learning (ML) models and highlights the importance of accessible tools for clinicians, enabling them to develop AI solutions and make advanced diagnostic tools available even in resource-limited settings. The focus is on differentiating normal/reactive cells from neoplastic cells in pleural effusions linked to lung adenocarcinoma. <b>Methods:</b> A dataset from the Cytopathology Unit at the Sant’Andrea University Hospital comprising 969 raw images, annotated with 3130 single mesothelial cells and 3260 adenocarcinoma cells, was categorized into two classes based on morphological features. Object-detection models were developed using YOLOv8 and the latest YOLOv11 instance segmentation models. <b>Results:</b> The models achieved an Intersection over Union (IoU) score of 0.72, demonstrating robust performance in class prediction for both categories, with YOLOv11 showing performance improvements over YOLOv8 in different metrics. <b>Conclusions:</b> The application of machine learning in cytopathology offers clinicians valuable support in differential diagnosis while also expanding their ability to engage with AI tools and methodologies. The diagnosis of MPEs is marked by substantial morphological and technical variability, underscoring the need for high-quality datasets and advanced deep-learning models. These technologies have the potential to enhance data interpretation and support more effective clinical treatment strategies in the era of precision medicine.https://www.mdpi.com/2075-4418/15/10/1240cytopathologycytologypleural effusionlung adenocarcinomamachine learningYOLOv8
spellingShingle Enrico Giarnieri
Elisabetta Carico
Stefania Scarpino
Alberto Ricci
Pierdonato Bruno
Simone Scardapane
Daniele Giansanti
Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models
Diagnostics
cytopathology
cytology
pleural effusion
lung adenocarcinoma
machine learning
YOLOv8
title Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models
title_full Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models
title_fullStr Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models
title_full_unstemmed Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models
title_short Bringing AI to Clinicians: Simplifying Pleural Effusion Cytology Diagnosis with User-Friendly Models
title_sort bringing ai to clinicians simplifying pleural effusion cytology diagnosis with user friendly models
topic cytopathology
cytology
pleural effusion
lung adenocarcinoma
machine learning
YOLOv8
url https://www.mdpi.com/2075-4418/15/10/1240
work_keys_str_mv AT enricogiarnieri bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels
AT elisabettacarico bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels
AT stefaniascarpino bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels
AT albertoricci bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels
AT pierdonatobruno bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels
AT simonescardapane bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels
AT danielegiansanti bringingaitoclinicianssimplifyingpleuraleffusioncytologydiagnosiswithuserfriendlymodels