BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection

Ming Kang; Chee-Ming Ting; Fung Fung Ting; Raphaël C.W. Phan

doi:10.1007/978-3-031-72111-3_4

BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection

Ming Kang, Chee-Ming Ting, Fung Fung Ting, Raphaël C.W. Phan

Malaysia Sch of Information Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

6 Citations (Scopus)

Abstract

You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP₅₀ compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.

Original language	English
Title of host publication	27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII
Editors	Marius George Linguraru, Qi Dou, Aasa Feragen, Stamatia Giannarou, Ben Glocker, Karim Lekadir, Julia A. Schnabel
Place of Publication	Cham Switzerland
Publisher	Springer
Pages	35-45
Number of pages	11
ISBN (Electronic)	9783031721113
ISBN (Print)	9783031721106
DOIs	https://doi.org/10.1007/978-3-031-72111-3_4
Publication status	Published - 2024
Event	Medical Image Computing and Computer-Assisted Intervention 2024 - Marrakesh, Morocco Duration: 6 Oct 2024 → 10 Oct 2024 Conference number: 27th https://link.springer.com/book/10.1007/978-3-031-72117-5 (Proceedings) https://conferences.miccai.org/2024/en/ (Website)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	15008
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	Medical Image Computing and Computer-Assisted Intervention 2024
Abbreviated title	MICCAI 2024
Country/Territory	Morocco
City	Marrakesh
Period	6/10/24 → 10/10/24
Internet address	https://link.springer.com/book/10.1007/978-3-031-72117-5 (Proceedings) https://conferences.miccai.org/2024/en/ (Website)

Keywords

attention mechanism
feature fusion
lesion detection
Medical image detection
YOLO

Access to Document

10.1007/978-3-031-72111-3_4

Cite this

Kang, M., Ting, C.-M., Ting, F. F., & Phan, R. C. W. (2024). BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection. In M. G. Linguraru, Q. Dou, A. Feragen, S. Giannarou, B. Glocker, K. Lekadir, & J. A. Schnabel (Eds.), 27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII (pp. 35-45). (Lecture Notes in Computer Science; Vol. 15008). Springer. https://doi.org/10.1007/978-3-031-72111-3_4

Kang, Ming ; Ting, Chee-Ming ; Ting, Fung Fung et al. / BGF-YOLO : Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection. 27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII. editor / Marius George Linguraru ; Qi Dou ; Aasa Feragen ; Stamatia Giannarou ; Ben Glocker ; Karim Lekadir ; Julia A. Schnabel. Cham Switzerland : Springer, 2024. pp. 35-45 (Lecture Notes in Computer Science).

@inproceedings{a973bda583a64bee8b494333d8bad020,

title = "BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection",

abstract = "You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP50 compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.",

keywords = "attention mechanism, feature fusion, lesion detection, Medical image detection, YOLO",

author = "Ming Kang and Chee-Ming Ting and Ting, {Fung Fung} and Phan, {Rapha{\"e}l C.W.}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.; Medical Image Computing and Computer-Assisted Intervention 2024, MICCAI 2024 ; Conference date: 06-10-2024 Through 10-10-2024",

year = "2024",

doi = "10.1007/978-3-031-72111-3_4",

language = "English",

isbn = "9783031721106",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "35--45",

editor = "Linguraru, {Marius George} and Qi Dou and Aasa Feragen and Stamatia Giannarou and Ben Glocker and Karim Lekadir and Schnabel, {Julia A.}",

booktitle = "27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII",

url = "https://link.springer.com/book/10.1007/978-3-031-72117-5, https://conferences.miccai.org/2024/en/",

}

Kang, M, Ting, C-M , Ting, FF & Phan, RCW 2024, BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection. in MG Linguraru, Q Dou, A Feragen, S Giannarou, B Glocker, K Lekadir & JA Schnabel (eds), 27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII. Lecture Notes in Computer Science, vol. 15008, Springer, Cham Switzerland, pp. 35-45, Medical Image Computing and Computer-Assisted Intervention 2024, Marrakesh, Morocco, 6/10/24. https://doi.org/10.1007/978-3-031-72111-3_4

BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection. / Kang, Ming; Ting, Chee-Ming ; Ting, Fung Fung et al.
27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII. ed. / Marius George Linguraru; Qi Dou; Aasa Feragen; Stamatia Giannarou; Ben Glocker; Karim Lekadir; Julia A. Schnabel. Cham Switzerland: Springer, 2024. p. 35-45 (Lecture Notes in Computer Science; Vol. 15008).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - BGF-YOLO

T2 - Medical Image Computing and Computer-Assisted Intervention 2024

AU - Kang, Ming

AU - Ting, Chee-Ming

AU - Ting, Fung Fung

AU - Phan, Raphaël C.W.

N1 - Conference code: 27th

PY - 2024

Y1 - 2024

N2 - You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP50 compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.

AB - You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP50 compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.

KW - attention mechanism

KW - feature fusion

KW - lesion detection

KW - Medical image detection

KW - YOLO

UR - http://www.scopus.com/inward/record.url?scp=85206891922&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-72111-3_4

DO - 10.1007/978-3-031-72111-3_4

M3 - Conference Paper

AN - SCOPUS:85206891922

SN - 9783031721106

T3 - Lecture Notes in Computer Science

SP - 35

EP - 45

BT - 27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII

A2 - Linguraru, Marius George

A2 - Dou, Qi

A2 - Feragen, Aasa

A2 - Giannarou, Stamatia

A2 - Glocker, Ben

A2 - Lekadir, Karim

A2 - Schnabel, Julia A.

PB - Springer

CY - Cham Switzerland

Y2 - 6 October 2024 through 10 October 2024

ER -

Kang M, Ting CM , Ting FF , Phan RCW. BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection. In Linguraru MG, Dou Q, Feragen A, Giannarou S, Glocker B, Lekadir K, Schnabel JA, editors, 27th International Conference Marrakesh, Morocco, October 6–10, 2024 Proceedings, Part VIII. Cham Switzerland: Springer. 2024. p. 35-45. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-72111-3_4