Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations

Caner Özer; Patryk Rygiel; Bram de Wilde; Ilkay Öksüz; Jelmer M. Wolterink

doi:10.1007/978-3-032-09513-8_35

Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations

Caner Özer^*
, Patryk Rygiel
, Bram de Wilde
, Ilkay Öksüz
, Jelmer M. Wolterink

^*Corresponding author for this work

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

Abstract

Artifacts pose a significant challenge in medical imaging, impacting diagnostic accuracy and downstream analysis. While image-based approaches for detecting artifacts can be effective, they often rely on preprocessing methods that can lead to information loss and high-memory-demand medical images, thereby limiting the scalability of classification models. In this work, we propose the use of implicit neural representations (INRs) for image quality assessment. INRs provide a compact and continuous representation of medical images, naturally handling variations in resolution and image size while reducing memory overhead. We develop deep weight space networks, graph neural networks, and relational attention transformers that operate on INRs to achieve image quality assessment. Our method is evaluated on the ACDC dataset with synthetically generated artifact patterns, demonstrating its effectiveness in assessing image quality while achieving similar performance with fewer parameters.

Original language	English
Title of host publication	Machine Learning in Medical Imaging
Subtitle of host publication	16th International Workshop, MLMI 2025, Held in Conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings
Editors	Zhiming Cui, Islem Rekik, Heung-IL Suk, Xi Ouyang, Kaicong Sun, Sheng Wang
Place of Publication	Cham
Publisher	Springer
Pages	359-368
Number of pages	10
ISBN (Electronic)	978-3-032-09513-8
ISBN (Print)	978-3-032-09512-1
DOIs	https://doi.org/10.1007/978-3-032-09513-8_35
Publication status	Published - Jan 2026
Event	16th International Workshop on Machine Learning in Medical Imaging, MLMI 2025 - Daejeon, Korea, Republic of Duration: 23 Sept 2025 → 23 Sept 2025 Conference number: 16

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	16241
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Workshop

Workshop	16th International Workshop on Machine Learning in Medical Imaging, MLMI 2025
Abbreviated title	MLMI
Country/Territory	Korea, Republic of
City	Daejeon
Period	23/09/25 → 23/09/25

Keywords

2026 OA procedure
Implicit neural representations
Medical image quality assessment
Neural fields
Artifact detection

Access to Document

10.1007/978-3-032-09513-8_35

Cite this

Özer, C., Rygiel, P., de Wilde, B., Öksüz, I., & Wolterink, J. M. (2026). Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations. In Z. Cui, I. Rekik, H.-IL. Suk, X. Ouyang, K. Sun, & S. Wang (Eds.), Machine Learning in Medical Imaging: 16th International Workshop, MLMI 2025, Held in Conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings (pp. 359-368). (Lecture Notes in Computer Science; Vol. 16241). Springer. https://doi.org/10.1007/978-3-032-09513-8_35

Özer, Caner ; Rygiel, Patryk ; de Wilde, Bram et al. / Beyond Pixels : Medical Image Quality Assessment with Implicit Neural Representations. Machine Learning in Medical Imaging: 16th International Workshop, MLMI 2025, Held in Conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. editor / Zhiming Cui ; Islem Rekik ; Heung-IL Suk ; Xi Ouyang ; Kaicong Sun ; Sheng Wang. Cham : Springer, 2026. pp. 359-368 (Lecture Notes in Computer Science).

@inproceedings{a1489071e9284cdb8bf16aba97cc2d56,

title = "Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations",

abstract = "Artifacts pose a significant challenge in medical imaging, impacting diagnostic accuracy and downstream analysis. While image-based approaches for detecting artifacts can be effective, they often rely on preprocessing methods that can lead to information loss and high-memory-demand medical images, thereby limiting the scalability of classification models. In this work, we propose the use of implicit neural representations (INRs) for image quality assessment. INRs provide a compact and continuous representation of medical images, naturally handling variations in resolution and image size while reducing memory overhead. We develop deep weight space networks, graph neural networks, and relational attention transformers that operate on INRs to achieve image quality assessment. Our method is evaluated on the ACDC dataset with synthetically generated artifact patterns, demonstrating its effectiveness in assessing image quality while achieving similar performance with fewer parameters.",

keywords = "2026 OA procedure, Implicit neural representations, Medical image quality assessment, Neural fields, Artifact detection",

author = "Caner {\"O}zer and Patryk Rygiel and \{de Wilde\}, Bram and Ilkay {\"O}ks{\"u}z and Wolterink, \{Jelmer M.\}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.; 16th International Workshop on Machine Learning in Medical Imaging, MLMI 2025, MLMI ; Conference date: 23-09-2025 Through 23-09-2025",

year = "2026",

month = jan,

doi = "10.1007/978-3-032-09513-8\_35",

language = "English",

isbn = "978-3-032-09512-1",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "359--368",

editor = "Zhiming Cui and Islem Rekik and Heung-IL Suk and Xi Ouyang and Kaicong Sun and Sheng Wang",

booktitle = "Machine Learning in Medical Imaging",

address = "Germany",

}

Özer, C, Rygiel, P , de Wilde, B, Öksüz, I & Wolterink, JM 2026, Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations. in Z Cui, I Rekik, H-IL Suk, X Ouyang, K Sun & S Wang (eds), Machine Learning in Medical Imaging: 16th International Workshop, MLMI 2025, Held in Conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. Lecture Notes in Computer Science, vol. 16241, Springer, Cham, pp. 359-368, 16th International Workshop on Machine Learning in Medical Imaging, MLMI 2025, Daejeon, Korea, Republic of, 23/09/25. https://doi.org/10.1007/978-3-032-09513-8_35

Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations. / Özer, Caner; Rygiel, Patryk ; de Wilde, Bram et al.
Machine Learning in Medical Imaging: 16th International Workshop, MLMI 2025, Held in Conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. ed. / Zhiming Cui; Islem Rekik; Heung-IL Suk; Xi Ouyang; Kaicong Sun; Sheng Wang. Cham: Springer, 2026. p. 359-368 (Lecture Notes in Computer Science; Vol. 16241).

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

TY - GEN

T1 - Beyond Pixels

T2 - 16th International Workshop on Machine Learning in Medical Imaging, MLMI 2025

AU - Özer, Caner

AU - Rygiel, Patryk

AU - de Wilde, Bram

AU - Öksüz, Ilkay

AU - Wolterink, Jelmer M.

N1 - Conference code: 16

PY - 2026/1

Y1 - 2026/1

N2 - Artifacts pose a significant challenge in medical imaging, impacting diagnostic accuracy and downstream analysis. While image-based approaches for detecting artifacts can be effective, they often rely on preprocessing methods that can lead to information loss and high-memory-demand medical images, thereby limiting the scalability of classification models. In this work, we propose the use of implicit neural representations (INRs) for image quality assessment. INRs provide a compact and continuous representation of medical images, naturally handling variations in resolution and image size while reducing memory overhead. We develop deep weight space networks, graph neural networks, and relational attention transformers that operate on INRs to achieve image quality assessment. Our method is evaluated on the ACDC dataset with synthetically generated artifact patterns, demonstrating its effectiveness in assessing image quality while achieving similar performance with fewer parameters.

AB - Artifacts pose a significant challenge in medical imaging, impacting diagnostic accuracy and downstream analysis. While image-based approaches for detecting artifacts can be effective, they often rely on preprocessing methods that can lead to information loss and high-memory-demand medical images, thereby limiting the scalability of classification models. In this work, we propose the use of implicit neural representations (INRs) for image quality assessment. INRs provide a compact and continuous representation of medical images, naturally handling variations in resolution and image size while reducing memory overhead. We develop deep weight space networks, graph neural networks, and relational attention transformers that operate on INRs to achieve image quality assessment. Our method is evaluated on the ACDC dataset with synthetically generated artifact patterns, demonstrating its effectiveness in assessing image quality while achieving similar performance with fewer parameters.

KW - 2026 OA procedure

KW - Implicit neural representations

KW - Medical image quality assessment

KW - Neural fields

KW - Artifact detection

UR - https://www.scopus.com/pages/publications/105027568425

U2 - 10.1007/978-3-032-09513-8_35

DO - 10.1007/978-3-032-09513-8_35

M3 - Conference contribution

AN - SCOPUS:105027568425

SN - 978-3-032-09512-1

T3 - Lecture Notes in Computer Science

SP - 359

EP - 368

BT - Machine Learning in Medical Imaging

A2 - Cui, Zhiming

A2 - Rekik, Islem

A2 - Suk, Heung-IL

A2 - Ouyang, Xi

A2 - Sun, Kaicong

A2 - Wang, Sheng

PB - Springer

CY - Cham

Y2 - 23 September 2025 through 23 September 2025

ER -

Özer C, Rygiel P , de Wilde B, Öksüz I, Wolterink JM. Beyond Pixels: Medical Image Quality Assessment with Implicit Neural Representations. In Cui Z, Rekik I, Suk HIL, Ouyang X, Sun K, Wang S, editors, Machine Learning in Medical Imaging: 16th International Workshop, MLMI 2025, Held in Conjunction with MICCAI 2025, Daejeon, South Korea, September 23, 2025, Proceedings. Cham: Springer. 2026. p. 359-368. (Lecture Notes in Computer Science). doi: 10.1007/978-3-032-09513-8_35