Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis

Abstract Accurate detection and classification of cellular and non-cellular components in urine microscopy images are essential for early diagnosis of renal and systemic health conditions. This study presents an optimized object detection framework based on the Red Fox Optimization (RFO)-enabled Rob...

Full description

Saved in:
Bibliographic Details
Main Authors: Neeraj Dahiya, Deo Prakash, Shakti Kundu, Shanu Rakesh Kuttan, Isha Suwalka, Manel Ayadi, Mitiku Dubale, Arshad Hashmi
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-11725-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849235397379358720
author Neeraj Dahiya
Deo Prakash
Shakti Kundu
Shanu Rakesh Kuttan
Isha Suwalka
Manel Ayadi
Mitiku Dubale
Arshad Hashmi
author_facet Neeraj Dahiya
Deo Prakash
Shakti Kundu
Shanu Rakesh Kuttan
Isha Suwalka
Manel Ayadi
Mitiku Dubale
Arshad Hashmi
author_sort Neeraj Dahiya
collection DOAJ
description Abstract Accurate detection and classification of cellular and non-cellular components in urine microscopy images are essential for early diagnosis of renal and systemic health conditions. This study presents an optimized object detection framework based on the Red Fox Optimization (RFO)-enabled Roboflow-DEtection TRansformer (RF-DETR) model, designed to automate urine sediment analysis with high precision and low latency. The RF-DETR model leverages a transformer-based architecture with deformable attention and a DINOv2 (self-distillation with no labels) pre-trained visual backbone to capture multi-scale features effectively. RFO, a nature-inspired metaheuristic, is employed to fine-tune critical hyperparameters such as learning rate, decoder layers, and dropout, enhancing the model’s convergence and generalization capabilities. Experiments were conducted on the RF100-VL urine microscopy dataset, where the proposed model achieved a precision of 0.78, recall of 0.66, mAP@0.5 of 0.737, and mAP@0.5:0.95 of 0.44 after 100 training epochs. Compared to baseline models, the optimized RF-DETR demonstrated improved performance in detecting small and medium objects like leukocytes and erythrocytes—crucial components for urinary tract infection and kidney disease detection. The model’s NMS-free design and multi-resolution training enable real-time inference on both GPU and edge devices. Additionally, visualization tools such as confusion matrices, F1-curves, and prediction overlays validate the robustness and interpretability of the system. The results confirm the suitability of the RFO-optimized RF-DETR framework for clinical deployment, offering a powerful tool for automated, scalable, and accurate urine analysis. Future work will focus on lightweight model variants, enhanced small-object detection, and domain adaptation using self-supervised and vision-language learning techniques.
format Article
id doaj-art-ba4350bfb68f4869b4837814c6d0518f
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-ba4350bfb68f4869b4837814c6d0518f2025-08-20T04:02:46ZengNature PortfolioScientific Reports2045-23222025-07-0115112910.1038/s41598-025-11725-0Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosisNeeraj Dahiya0Deo Prakash1Shakti Kundu2Shanu Rakesh Kuttan3Isha Suwalka4Manel Ayadi5Mitiku Dubale6Arshad Hashmi7Department of Computer Science & Engineering, SRM UniversitySchool of Computer Science & Engineering, Faculty of Engineering, Shri Mata Vaishno Devi UniversityDepartment of Computer Science and Engineering, NIIT UniversityDepartment of Computer Science and Engineering, Chouksey Engineering CollegeDepartment of Research and Publication, Indira IVF Hospital LimitedDepartment of Information Systems, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman UniversityCollege of Natural and Computational Science, Gambella UniversityDepartment of Information Systems, Faculty of Computing and Information Technology in Rabigh (FCITR), King Abdulaziz UniversityAbstract Accurate detection and classification of cellular and non-cellular components in urine microscopy images are essential for early diagnosis of renal and systemic health conditions. This study presents an optimized object detection framework based on the Red Fox Optimization (RFO)-enabled Roboflow-DEtection TRansformer (RF-DETR) model, designed to automate urine sediment analysis with high precision and low latency. The RF-DETR model leverages a transformer-based architecture with deformable attention and a DINOv2 (self-distillation with no labels) pre-trained visual backbone to capture multi-scale features effectively. RFO, a nature-inspired metaheuristic, is employed to fine-tune critical hyperparameters such as learning rate, decoder layers, and dropout, enhancing the model’s convergence and generalization capabilities. Experiments were conducted on the RF100-VL urine microscopy dataset, where the proposed model achieved a precision of 0.78, recall of 0.66, mAP@0.5 of 0.737, and mAP@0.5:0.95 of 0.44 after 100 training epochs. Compared to baseline models, the optimized RF-DETR demonstrated improved performance in detecting small and medium objects like leukocytes and erythrocytes—crucial components for urinary tract infection and kidney disease detection. The model’s NMS-free design and multi-resolution training enable real-time inference on both GPU and edge devices. Additionally, visualization tools such as confusion matrices, F1-curves, and prediction overlays validate the robustness and interpretability of the system. The results confirm the suitability of the RFO-optimized RF-DETR framework for clinical deployment, offering a powerful tool for automated, scalable, and accurate urine analysis. Future work will focus on lightweight model variants, enhanced small-object detection, and domain adaptation using self-supervised and vision-language learning techniques.https://doi.org/10.1038/s41598-025-11725-0RF-DETR modelRed Fox optimization (RFO)Urine microscopyObject detectionHyperparameter tuningAutomated diagnosis
spellingShingle Neeraj Dahiya
Deo Prakash
Shakti Kundu
Shanu Rakesh Kuttan
Isha Suwalka
Manel Ayadi
Mitiku Dubale
Arshad Hashmi
Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis
Scientific Reports
RF-DETR model
Red Fox optimization (RFO)
Urine microscopy
Object detection
Hyperparameter tuning
Automated diagnosis
title Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis
title_full Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis
title_fullStr Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis
title_full_unstemmed Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis
title_short Optimised RFO tuned RF-DETR model for precision urine microscopy for renal and systemic disease diagnosis
title_sort optimised rfo tuned rf detr model for precision urine microscopy for renal and systemic disease diagnosis
topic RF-DETR model
Red Fox optimization (RFO)
Urine microscopy
Object detection
Hyperparameter tuning
Automated diagnosis
url https://doi.org/10.1038/s41598-025-11725-0
work_keys_str_mv AT neerajdahiya optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT deoprakash optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT shaktikundu optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT shanurakeshkuttan optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT ishasuwalka optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT manelayadi optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT mitikudubale optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis
AT arshadhashmi optimisedrfotunedrfdetrmodelforprecisionurinemicroscopyforrenalandsystemicdiseasediagnosis