EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Framework to Investigate the Effects of Observation Error on Neural Network Predictions of Fish Age

Derek W. Chamberlin, Thomas E. Helser, John D. Brogan, Irina M. Benson, Jason Conner, Andrew T. Chin, Christopher M. Gburski, Mary Elizabeth Matta, Julie A. Pearce, Kali R. Stone, Todd T. TenBrink, Morgan B. Arrington

Fish and Fisheries · 2026-07 · 27 (4) · 974-986

DOI: 10.1111/faf.70094 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Fish and Fisheries, online erschienen am 08.05.2026. Verfasst von Derek W. Chamberlin und 11 weiteren Autor:innen. Die Arbeit nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Alaska Fisheries Science Center, National Marine Fisheries Service.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Technological innovations for predicting fish age represent a paradigm shift from conventional age estimation methods used in fisheries science. Recently developed secondary methods rely on models trained on conventional age estimates, derived from subjective interpretation of growth patterns and a biological property of the fish to predict age. Hence, quantifying the error that propagates from these models and the conventional age estimates on which they are trained is critical to fully account for uncertainty in predicted ages used in fisheries stock assessments. We review the development of three secondary age prediction methods: image analysis, epigenetics and Fourier transform near‐infrared spectroscopy (FT‐NIRS); the use of artificial intelligence (AI) within each method; and present an approach to examine the effects of ageing error on AI age prediction models. As a case study, we conducted an empirical study, coupled with simulation, to quantify the effects of ageing error on the performance of a multimodal convolutional neural network (MMCNN) model used to predict eastern Bering Sea walleye pollock ( Gadus chalcogrammus ) ages from otoliths analysed using FT‐NIRS. Results indicated repeatability of predicted ages was high between instrument operators, while adding ageing error resulted in a slight decrease in model performance from R 2 = 0.92 and CV = 7.6% to R 2 = 0.87 and CV = 10.2% on test datasets. Our results also suggest the MMCNN model is robust to noise in calibration age data and model performance may be better than performance metrics indicate when the model is evaluated against data with increased error.

Förderung

  • Alaska Fisheries Science Center
  • National Marine Fisheries Service

Zitieren

APA

Chamberlin, D. W., Helser, T. E., Brogan, J. D., Benson, I. M., Conner, J., Chin, A. T., Gburski, C. M., Matta, M. E., Pearce, J. A., Stone, K. R., TenBrink, T. T., & Arrington, M. B. (2026). A Framework to Investigate the Effects of Observation Error on Neural Network Predictions of Fish Age. Fish and Fisheries, 27(4), 974-986. https://doi.org/10.1111/faf.70094

Vancouver

Chamberlin DW, Helser TE, Brogan JD, Benson IM, Conner J, Chin AT, et al. A Framework to Investigate the Effects of Observation Error on Neural Network Predictions of Fish Age. Fish and Fisheries. 2026;27(4):974-986. doi:10.1111/faf.70094

BibTeX
@article{Chamberlin2026,
  author = {Chamberlin, Derek W. and Helser, Thomas E. and Brogan, John D. and Benson, Irina M. and Conner, Jason and Chin, Andrew T. and Gburski, Christopher M. and Matta, Mary Elizabeth and Pearce, Julie A. and Stone, Kali R. and TenBrink, Todd T. and Arrington, Morgan B.},
  title = {A Framework to Investigate the Effects of Observation Error on Neural Network Predictions of Fish Age},
  journal = {Fish and Fisheries},
  year = {2026},
  volume = {27},
  number = {4},
  pages = {974--986},
  doi = {10.1111/faf.70094}
}
Kostenlos starten →