Unique Deep Radiomic Signature Shows NMN Treatment Reverses Morphology of Oocytes from Aged Mice

Abbas Habibalahi; Jared M Campbell; Michael J Bertoldo; Saabah B Mahbub; Dale M Goss; William L Ledger; Robert B Gilchrist; Lindsay E Wu; Ewa M Goldys

doi:10.3390/biomedicines10071544

Unique Deep Radiomic Signature Shows NMN Treatment Reverses Morphology of Oocytes from Aged Mice

Biomedicines. 2022 Jun 29;10(7):1544. doi: 10.3390/biomedicines10071544.

Authors

Abbas Habibalahi¹, Jared M Campbell¹, Michael J Bertoldo^{2

3}, Saabah B Mahbub¹, Dale M Goss³, William L Ledger², Robert B Gilchrist², Lindsay E Wu³, Ewa M Goldys¹

Affiliations

¹ ARC Centre of Excellence Centre for Nanoscale Biophotonics, University of New South Wales Sydney, Sydney, NSW 2052, Australia.
² Fertility & Research Centre, School of Clinical Medicine, University of New South Wales Sydney, Sydney, NSW 2052, Australia.
³ School of Medical Sciences, University of New South Wales Sydney, Sydney, NSW 2052, Australia.

Abstract

The purpose of this study is to develop a deep radiomic signature based on an artificial intelligence (AI) model. This radiomic signature identifies oocyte morphological changes corresponding to reproductive aging in bright field images captured by optical light microscopy. Oocytes were collected from three mice groups: young (4- to 5-week-old) C57BL/6J female mice, aged (12-month-old) mice, and aged mice treated with the NAD+ precursor nicotinamide mononucleotide (NMN), a treatment recently shown to rejuvenate aspects of fertility in aged mice. We applied deep learning, swarm intelligence, and discriminative analysis to images of mouse oocytes taken by bright field microscopy to identify a highly informative deep radiomic signature (DRS) of oocyte morphology. Predictive DRS accuracy was determined by evaluating sensitivity, specificity, and cross-validation, and was visualized using scatter plots of the data associated with three groups: Young, old and Old + NMN. DRS could successfully distinguish morphological changes in oocytes associated with maternal age with 92% accuracy (AUC~1), reflecting this decline in oocyte quality. We then employed the DRS to evaluate the impact of the treatment of reproductively aged mice with NMN. The DRS signature classified 60% of oocytes from NMN-treated aged mice as having a 'young' morphology. In conclusion, the DRS signature developed in this study was successfully able to detect aging-related oocyte morphological changes. The significance of our approach is that DRS applied to bright field oocyte images will allow us to distinguish and select oocytes originally affected by reproductive aging and whose quality has been successfully restored by the NMN therapy.

Keywords: NMN; aging; machine learning; morphology; oocyte.

Grants and funding

This work is partially supported by the Australian Research Council: CE140100003, and DP170101863 and by National Health and Medical Research Council: APP1139763, APP1117538, APP1127821, APP1122484, APP1103689, and APP1093643. The salaries of M.J.B. and D.M.G. working in the lab of L.E.W. and R.B.G. were partly paid by contract research from Jumpstart Fertility to UNSW. We acknowledge NSW cancer Institute for early career fellowship (ECF1291).