Physics-Informed Unrolled Learning for ATR-FTIR Spectral Unmixing
- 1Department of Mathematics, King Abdulaziz University, Jeddah 21589, Saudi Arabia
Abstract
Quantitative infrared spectroscopy of bone reads chemistry from ratios of band intensities, but attenuated total reflectance distorts those intensities before they are measured. We treat the distortion as a model to be inverted rather than a nuisance to be filtered. A causal Lorentz parameterisation of the dielectric function, an effective-medium mixture, the Fresnel relations for a finite contact crystal and the instrument line shape are composed into one differentiable operator, and its Jacobian is derived in closed form. Three consequences are quantified. Anomalous dispersion displaces the apparent phosphate maximum by 7 to 36 wavenumbers depending on packing, so band position thresholds do not transfer between laboratories. The displacement varies with composition, so the I1030/I1020 crystallinity ratio responds 1.93 times more strongly to mineral content at fixed crystallinity than to the entire plausible crystallinity range. Absorbance is not affine in composition, which invalidates bilinear unmixing. We prove that closure and structural anticorrelation force the sufficiently scattered condition to fail for bone, and replace it with spectral separability, a checkable condition on band spacing that yields a local identifiability theorem with an explicit lower bound on the smallest singular value. Inference uses a Levenberg–Marquardt iteration unrolled into a network with five learned scalars per layer, for which we prove local linear convergence under a per-layer contraction condition, for constant and for layer-dependent schedules, and a Rademacher bound of order TlogT/n that is conditional on non-expansive layers; on the trained networks the median contraction is below one at every layer, but non-expansiveness does not hold. On a generator whose band ratios all lie inside published ranges, the estimator reaches the Cramér–Rao bound to within 3 to 6 per cent across a sixteenfold range of noise, recovers volume fractions 143 times more accurately than multivariate curve resolution, and outperforms a network with 1.1×105 parameters by three orders of magnitude at twenty-four training spectra. These figures hold when the data follow the operator. Under deliberate mismatch in band positions, line shape, baseline, effective-medium law, contact physics and noise, the advantage over curve resolution persists at factors of 2 to 108, but the loss rises to many times the bound and the nominal intervals cover the truth in as few as 5 per cent of cases, so sampling uncertainty is reported separately from systematic model uncertainty. The same information matrix prices the experiment: contact variability inflates the standard error on composition by 3.6 while leaving the band parameters untouched, and an optimal choice of crystal and geometry recovers 23.6 nats under signal-independent noise, falling to between 0.6 and 17.9 nats once throughput loss and detector-limited noise are included.
Related
- Physics-Informed Explainable Machine Learning for Hybrid State of Health and Remaining Useful Life Estimation of Lithium-Ion Batteries
- A systematic comparison of hard- and soft-constrained physics-informed molecular machine learning with the Clapeyron equation
- Physics-Informed Deep Learning Surrogate for Mechanical-Response Prediction and Failure Assessment of High-Performance Polymer Composites
- Thermodynamically Constrained Deep Learning for Hydrogen Sorption Prediction: Overcoming Classical Isotherm Generalization Failure through Physics-Informed Cross-Lithology Transferability
- Physics-informed impedance spectroscopy for diagnosing XLPE cable insulation aging based on a second-order Havriliak-Negami dielectric model
BibTeX
@article{alofi2026physicsinformed,
title = {{Physics-Informed Unrolled Learning for ATR-FTIR Spectral Unmixing}},
author = {Abdulaziz S. Alofi and Abdullah Al Mazrooei},
journal = {Mathematics},
year = {2026},
month = oct,
doi = {10.3390/math14193619},
url = {https://www.mdpi.com/2227-7390/14/19/3619},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Physics-Informed Unrolled Learning for ATR-FTIR Spectral Unmixing
AU - Abdulaziz S. Alofi
AU - Abdullah Al Mazrooei
PY - 2026
DA - 2026/10/06
JO - Mathematics
DO - 10.3390/math14193619
UR - https://www.mdpi.com/2227-7390/14/19/3619
AB - Quantitative infrared spectroscopy of bone reads chemistry from ratios of band intensities, but attenuated total reflectance distorts those intensities before they are measured. We treat the distortion as a model to be inverted rather than a nuisance to be filtered. A causal Lorentz parameterisation of the dielectric function, an effective-medium mixture, the Fresnel relations for a finite contact crystal and the instrument line shape are composed into one differentiable operator, and its Jacobian is derived in closed form. Three consequences are quantified. Anomalous dispersion displaces the apparent phosphate maximum by 7 to 36 wavenumbers depending on packing, so band position thresholds do not transfer between laboratories. The displacement varies with composition, so the I1030/I1020 crystallinity ratio responds 1.93 times more strongly to mineral content at fixed crystallinity than to the entire plausible crystallinity range. Absorbance is not affine in composition, which invalidates bilinear unmixing. We prove that closure and structural anticorrelation force the sufficiently scattered condition to fail for bone, and replace it with spectral separability, a checkable condition on band spacing that yields a local identifiability theorem with an explicit lower bound on the smallest singular value. Inference uses a Levenberg–Marquardt iteration unrolled into a network with five learned scalars per layer, for which we prove local linear convergence under a per-layer contraction condition, for constant and for layer-dependent schedules, and a Rademacher bound of order TlogT/n that is conditional on non-expansive layers; on the trained networks the median contraction is below one at every layer, but non-expansiveness does not hold. On a generator whose band ratios all lie inside published ranges, the estimator reaches the Cramér–Rao bound to within 3 to 6 per cent across a sixteenfold range of noise, recovers volume fractions 143 times more accurately than multivariate curve resolution, and outperforms a network with 1.1×105 parameters by three orders of magnitude at twenty-four training spectra. These figures hold when the data follow the operator. Under deliberate mismatch in band positions, line shape, baseline, effective-medium law, contact physics and noise, the advantage over curve resolution persists at factors of 2 to 108, but the loss rises to many times the bound and the nominal intervals cover the truth in as few as 5 per cent of cases, so sampling uncertainty is reported separately from systematic model uncertainty. The same information matrix prices the experiment: contact variability inflates the standard error on composition by 3.6 while leaving the band parameters untouched, and an optimal choice of crystal and geometry recovers 23.6 nats under signal-independent noise, falling to between 0.6 and 17.9 nats once throughput loss and detector-limited noise are included.
SN - 2227-7390
ER -