In this talk I will describe the use of a range of advanced photonics-based approaches of light sheet microscopy and digital holographic microscopy for label-free imaging. A driver for this work is the understanding of the development of the pre-implantation mammalian embryo [1-4] and improve IVF outcomes.
Embryo quality is a crucial factor affecting live birth outcomes. However, an accurate diagnostic for embryo quality remains elusive in the IVF clinic. Exploiting advanced optical imaging can assess the embryo in 3D and determine its metabolic rate and other physical parameters. This may ultimately prove to be a new multimodal diagnostic approach for embryo health.
Cellular metabolism is a key regulator of energetics, cell growth, regeneration, and homeostasis. The endogenous metabolic cofactors, nicotinamide adenine dinucleotide (phosphate) (NAD(P)H) and flavin adenine dinucleotide (FAD) can be imaged through their autofluorescence. By performing this with hyperspectral imaging at subcellular resolution may assist in determining embryo viability in a clinical setting. Such hyperspectral imaging can be used to determine the ploidy status of the embryo [1]. By using new implementations of light sheet imaging, we can extend this imaging to 3D [2]. Separately, we can tailor digital holographic microscopy (DHM) to measure spatio-temporal changes in refractive index during the development of the embryo that are reflective of its lipid content. Accumulation of intracellular lipid is known to compromise embryo health thus making this a further useful approach for diagnosis [3]. Overall, advanced photonics adds useful, label-free multimodal information for IVF success and can be gentle enough to not effect viability [4].
[1] T. C. Y. Tan et al., Hum. Reprod. 37(1), 14–29 (2021).
[2] Josephine Morizet et al., ACS Photonics 10, 4177-4187 (2023)
[3] George O. Dwapanyin et al., Biomed. Opt. Express 14, 3327-3342 (2023)
[4] C. A. Campugan et al., J. Assist. Reprod. Genet. 39(8), 1825–1837 (2022).
Recent strides in machine learning have shown that computation can be performed by practically any controllable physical system that responds to physical stimuli encoding data [1]. This perspective opens new frontiers for computational approaches using Physical Neural Networks (PNNs) and provides a framework to deepen our understanding of their biological counterparts—neural circuits in living organisms. To fully leverage this potential, PNNs must be trained with a nuanced awareness of the physical nature of signal and noise, where signal is defined relative to the specific computational task. This perspective aligns closely with approaches to determining the fundamental limits to sensing as set by quantum theory but extends these ideas to a new level to encompass broader computational opportunities. I will share insights from our approach to this emerging domain of inquiry and highlight some recent results [2, 3, 4].
Based on work with Fangjun Hu, Saeed A. Khan, Gerasimos Angelatos, Marti Vives, Esin Türeci, Graham E. Rowlands, Guilhem J. Ribeill, Nicholas Bronn.
[1] Aspen Center for Physics Winter Conference, Computing with Physical Systems, https://computingwithphysicalsystems.com/2024/
[2] F. Hu et al. `Tackling Sampling Noise in Physical Systems for Machine Learning Applications: Fundamental Limits and Eigentasks." Phys. Rev. X 13, 041020 (2023).
[3] S. A. Khan et al., `A neural processing approach to quantum state discrimination", arxiv:2409.03748.
[4] F. Hu et al. `Overcoming the Coherence Time Barrier in Quantum Machine Learning on Temporal Data", Nature Commun. 15, 7491 (2024).
The study of many-body quantum Hamiltonians presents significant theoretical and experimental challenges because their complexity. While quantum simulators provide powerful experimental platforms for probing these systems, their effectiveness depends on precise control and efficient approximations. A major obstacle arises in simulating high-body-interaction systems — such as lattice gauge theories or topological codes — given the limitations of experimental systems. A way to overcome such difficulties is to search how to simulate a complex Hamiltonians with a simpler ones which could be experimentally feasible, however, is that possible? Here, I want to discuss how to use tools from quantum information to optimize simulation accuracy. Finally, we also address the geometric structure of the local Hamiltonian manifold and link such structure with the convex sets of quantum states. By unifying all these perspectives, we aim to advance both theoretical understanding and practical implementations of quantum simulations.
Volumes of parent Hamiltonians for benchmarking quantum simulators
M García- Díaz, G Sentís, RM Tapia, A Sanpera, Physical Review Research, 2023
Simplifying the simulation of local Hamiltonian dynamics
A Usui, A Sanpera, M García-Díaz - Physical Review Research, 2024
Embryogenesis, the development of a complex multicellular organism
from a single fertilized egg cell, is a fascinating self-organization
phenomenon that invokes biochemical and physical cues on multiple
length and time scales. To study the role of physics during early
developmental stages, the small nematode Caenorhabditis elegans is
an excellent model organism because of its simplicity, transparency,
genetic accessibility, and developmental reproducibility.
Using light-sheet microscopy, we have explored how physical cues
determine cell arrangement, cell division timing, the size-adaption
of nuclei, and the diffusive transport of condensate-like protein
granules in early embryonic stages of C. elegans. Our results reveal
that granules mostly perform non-Markovian random walks while the
interplay of limiting components and mechanical forces create simple
and robust physical cues that allow the embryo to self-organize and
develop on autopilot until gastrulation.
