pith. sign in

arxiv: 2605.05276 · v1 · submitted 2026-05-06 · 💻 cs.IT · math.IT· math.OC· math.PR

On Unbiased Parameter Estimation and Signal Reconstruction

Joonas Lahtinen This is my paper

Pith reviewed 2026-05-08 16:19 UTC · model grok-4.3

classification 💻 cs.IT math.ITmath.OCmath.PR
keywords unbiased parameter estimationsignal reconstructionexact reconstructibilitynoise robustnessprobability measuremagnitude orderingsensor SNR trade-off
0
0 comments X

The pith

Unbiased parameter estimation derives upper bounds on recoverable non-zero parameters in the noiseless case while explaining noise robustness via a probability measure on magnitude ordering.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper extends depth-unbiased source localization to unbiased parameter estimation and signal reconstruction for an arbitrary number of non-zero parameters. It derives upper bounds on the number of parameters that can be recovered exactly when noise is absent, drawing on exact reconstructibility ideas. A probability measure is introduced to assess the chance of recovering all non-zero parameters with their magnitudes in the correct order. These elements together supply a mathematical account of why standardized unbiased methods exhibit noise robustness. The analysis further identifies a direct trade-off in which the number of sensors compensates for lower signal-to-noise ratios.

Core claim

Generalizing unbiased estimation beyond source localization produces upper bounds on the number of recoverable non-zero parameters that follow from exact reconstructibility in the noiseless case. A probability measure is defined to quantify the success of obtaining every non-zero parameter with correct magnitude order, which accounts for the noise robustness of unbiased approaches and exposes a trade-off between sensor count and required signal-to-noise ratio.

What carries the argument

Exact reconstructibility applied to the generalized unbiased estimation setting, which supplies the upper bounds on parameter count and the probability measure for ordered magnitude recovery.

If this is right

  • In the complete absence of noise only a bounded number of parameters admit exact unbiased recovery.
  • The probability of correct magnitude ordering rises under the conditions that produce the observed robustness to added noise.
  • Increasing the number of sensors directly lowers the signal-to-noise ratio needed to maintain reliable recovery.
  • Numerical experiments confirm that the theoretical bounds and the sensor-SNR trade-off hold in practice.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The probability measure could be used to forecast recovery reliability in sensor-array designs before deployment.
  • The identified trade-off suggests that hardware cost models could be optimized by balancing sensor quantity against expected operating noise levels.
  • The same bounding approach might be tested on estimation tasks outside linear signal models to check whether the reconstructibility limits persist.

Load-bearing premise

That exact reconstructibility concepts from compressed sensing apply directly to define bounds and robustness properties in this generalized unbiased estimation framework.

What would settle it

A noiseless simulation or calculation in which the unbiased estimator recovers more non-zero parameters than the derived upper bound while preserving correct magnitudes would falsify the bound.

Figures

Figures reproduced from arXiv: 2605.05276 by Joonas Lahtinen.

Figure 1
Figure 1. Figure 1: Reconstructions of noisy Shepp-Logan phantoms from 22 radial samples and the highlights of the errors in the nebula color map plotted over the clean phantom. Reconstructions are computed with Spatial Total Variation Split Bregman (STVSB), Unbiased Gaussian Estimate (UGE), and Exact Coloring Bayesian method EC Bayesian, which has the prior information about the ratio of all seven colors appearing in the tru… view at source ↗
Figure 2
Figure 2. Figure 2: Relative errors of the method for each foremost iteration. Iterations are presented on a logarithmic x-axis and errors on a y-axis. STVSB (light blue) is computed over 100 split Bregman iterations. UGE (red) and EC Bayesian (violet) are computed over 500 samples. the noise is correlated, indicating that the STVSB reconstruction is essentially a reconstruction of the noisy image. In contrast, Bayesian metho… view at source ↗
Figure 3
Figure 3. Figure 3: The figure follows the same logic as view at source ↗
Figure 4
Figure 4. Figure 4: Relative errors of the method for each foremost iteration. Iterations are presented on a logarithmic x-axis and errors on the y-axis. STVSB (light blue) is computed over 100 split Bregman iterations. UGE (red) and EC Bayesian (violet) are computed over 1,000 samples in 10 % Gaussian i.i.d. noise case and otherwise limited to 500 samples due to the convergence. 5 % noise 0.25 0.25 0.5 0.5 0.5 0.75 0.75 0.9 … view at source ↗
Figure 5
Figure 5. Figure 5: Probability contour map to reconstruct exactly the source in the conductivity disk without knowledge about the observation noise, where the sensors lie only on one half of the disk (gray circles on the upper half). however, that inference requires human input. Orientation estimate is the best with sLORETA2D. In scenario C, most methods indicate two sources near each other, with the more superficial one clo… view at source ↗
Figure 6
Figure 6. Figure 6: Reconstruction probability under different levels of signal-to-noise ratio as defined in Eq. (31). Probability curves are plotted for four different number of sensors, 16, 32, 64, 128, as displayed in the legend. Top graph shows the probabilities for a model, where the source is described by three parameters, e.g., sources have free orientation in a three-dimensional space, and bottom graph shows the same … view at source ↗
Figure 7
Figure 7. Figure 7: Dipolar source estimations of MNE, sLORETA, sLORETA2D, and Unbiased Bayesian method (UGE) in the homogeneous conductivity disk in three different source scenarios A, B, and C. The red cross indicates the location of the true source, and the red arrow represents the 2-dimensional source orientation. Locations of the source estimates are represented by colored circles and orientations by the arrows of the sa… view at source ↗
Figure 8
Figure 8. Figure 8: The distributed source estimate of MNE and Z-score distribution of standardized methods and unbiased Gaussian estimate (UGE) displayed as contours on the conductivity disk for three different source scenarios. The true source is depicted by a red cross and arrow view at source ↗
Figure 9
Figure 9. Figure 9: Z-score distributions of sLORETA, sLORETA3D, and UGE interpolated over the MRI slices, where the superficial near-field source and deep far-field source are located in subsequent image rows. The turquoise circle indicates the location of the corresponding source view at source ↗
read the original abstract

In this paper, we expand the theory of depth-unbiased source localization to unbiased parameter estimation and signal reconstruction of an arbitrary number of non-zero parameters to be recovered. The topic touches on the concept of exact reconstructibility, most commonly known in compressed sensing and multisource estimation in various imaging problems. The theoretical results derive upper bounds on the number of recoverable parameters in the noiseless case, and a probability measure is defined to assess the probability of obtaining all non-zero parameters with correct magnitude order. The work provides a mathematical explanation of the open question regarding the noise robustness of standardized and unbiased methods. Also, the paper reveals a trade-off between the number of sensors and the signal-to-noise ratio. Numerical experiments demonstrate the theoretical findings.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper extends the theory of depth-unbiased source localization to unbiased parameter estimation and signal reconstruction for an arbitrary number of non-zero parameters. It claims to derive upper bounds on the number of recoverable parameters in the noiseless case, define a probability measure for recovering all non-zero parameters with correct magnitude order, provide a mathematical explanation of the noise robustness of standardized and unbiased methods, identify a trade-off between the number of sensors and SNR, and validate the results via numerical experiments. The work connects to concepts of exact reconstructibility from compressed sensing.

Significance. If the claimed derivations and explanations hold, the results could help clarify limits on unbiased recovery and noise robustness in generalized parameter estimation settings, potentially informing applications in imaging and multisource problems. The introduction of a probability measure and the sensor-SNR trade-off would be useful additions if rigorously supported. However, the absence of explicit derivations, error analysis, and experimental details in the presentation limits the immediate impact.

major comments (2)
  1. Abstract: the manuscript states that upper bounds on recoverable parameters are derived in the noiseless case and that a probability measure is defined, yet no equations, proofs, or explicit bounds are supplied. This is load-bearing for the central claim, as the theoretical contribution cannot be evaluated without the derivations or the mapping from compressed-sensing reconstructibility to the unbiased estimation setting.
  2. Abstract: numerical experiments are said to demonstrate the theoretical findings, but no details on setup, parameter values, data-exclusion rules, or quantitative results (e.g., error metrics or success rates) are provided. This prevents verification of whether the experiments support the bounds, probability measure, or noise-robustness claims.
minor comments (1)
  1. Abstract: the phrase 'the open question regarding the noise robustness' is used without stating what the open question is or citing prior literature, reducing clarity for readers unfamiliar with the specific gap being addressed.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and will revise the manuscript accordingly to improve clarity and completeness.

read point-by-point responses

  1. Referee: Abstract: the manuscript states that upper bounds on recoverable parameters are derived in the noiseless case and that a probability measure is defined, yet no equations, proofs, or explicit bounds are supplied. This is load-bearing for the central claim, as the theoretical contribution cannot be evaluated without the derivations or the mapping from compressed-sensing reconstructibility to the unbiased estimation setting.

    Authors: The upper bounds on the number of recoverable non-zero parameters in the noiseless case and the probability measure for recovering all parameters with correct magnitude order are derived in the main text, with the connection to exact reconstructibility from compressed sensing explained in the introduction and theoretical development sections. We agree the abstract should be more self-contained. In revision we will insert the explicit upper-bound expression, the definition of the probability measure, and a brief reference to the proof strategy and compressed-sensing mapping so that the central claims can be evaluated directly from the abstract. revision: yes

  2. Referee: Abstract: numerical experiments are said to demonstrate the theoretical findings, but no details on setup, parameter values, data-exclusion rules, or quantitative results (e.g., error metrics or success rates) are provided. This prevents verification of whether the experiments support the bounds, probability measure, or noise-robustness claims.

    Authors: The numerical experiments validating the bounds, probability measure, noise robustness, and sensor-SNR trade-off are presented in the experimental section, including Monte Carlo trials and quantitative metrics. We will revise the abstract to summarize the key setup parameters (number of sensors, SNR range, trial count), data-handling rules, and main quantitative outcomes (success rates for magnitude-order recovery and error metrics). All supporting details will remain fully documented in the main text for reproducibility. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The paper expands depth-unbiased source localization to general unbiased parameter estimation, deriving upper bounds on recoverable parameters in the noiseless case and defining a probability measure for correct magnitude ordering. These results are framed as theoretical extensions of exact reconstructibility concepts from compressed sensing. No equations or steps in the abstract or description reduce a claimed prediction or bound to a fitted input by construction, nor do they rely on self-citation chains or ansatzes smuggled from prior work by the same author. The central claims rest on independent mathematical derivations applied to the generalized setting, with numerical experiments serving as validation rather than input. This is the common honest outcome for a theoretical paper whose load-bearing steps are not shown to collapse into their own definitions.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no explicit free parameters, axioms, or invented entities can be identified from the provided text.

pith-pipeline@v0.9.0 · 5414 in / 1041 out tokens · 31315 ms · 2026-05-08T16:19:06.174876+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

36 extracted references · 36 canonical work pages

  1. [1]

    Review on solving the inverse problem in EEG source analysis.Journal of neuroengineering and rehabilitation, 5(1):25–25, 2008

    Roberta Grech, Tracey Cassar, Joseph Muscat, Kenneth P Camilleri, Simon G Fabri, Michalis Zervakis, Petros Xanthopoulos, Vangelis Sakkalis, and Bart Vanrumste. Review on solving the inverse problem in EEG source analysis.Journal of neuroengineering and rehabilitation, 5(1):25–25, 2008

    work page 2008

  2. [2]

    Conditionally Gaussian Hypermodels for Cerebral Source Localization.SIAM Journal on Imaging Sciences, 2(3):879– 909, 2009

    Daniela Calvetti, Harri Hakula, Sampsa Pursiainen, and Erkki Somersalo. Conditionally Gaussian Hypermodels for Cerebral Source Localization.SIAM Journal on Imaging Sciences, 2(3):879– 909, 2009

    work page 2009

  3. [3]

    Gonzalez, and Fabio Babiloni

    Rolando Grave de Peralta, Olaf Hauk, Sara L. Gonzalez, and Fabio Babiloni. The neuroelectromagnetic inverse problem and the zero dipole localization error.Computational Intelligence and Neuroscience, 2009(1):137–147, 2009

    work page 2009

  4. [4]

    Solving the Electrocardiography Inverse Problem by Using an Optimal Algorithm Based on the Total Least Squares Theory

    Guofa Shou, Ling Xia, and Mingfeng Jiang. Solving the Electrocardiography Inverse Problem by Using an Optimal Algorithm Based on the Total Least Squares Theory. InThird International Conference on Natural Computation (ICNC 2007), volume 5, pages 115–119. IEEE, 2007

    work page 2007

  5. [5]

    Jes s Carrera, Andr s Alcolea, Agust n Medina, Juan Hidalgo, and Luit J. Slooten. Inverse problem in hydrogeology.Hydrogeology journal, 13(1):206–222, 2005

    work page 2005

  6. [6]

    T.C Johnson and D Wellman. Accurate modelling and inversion of electrical resistivity data in the presence of metallic infrastructure with known location and dimension.Geophysical journal international, 202(2):1096–1108, 2015

    work page 2015

  7. [7]

    Progress of electrical resistance tomography application in oil and gas reservoirs for development dynamic monitoring.Processes, 11(10):2950–, 2023

    Wenyang Shi, Guangzhi Yin, Mi Wang, Lei Tao, Mengjun Wu, Zhihao Yang, Jiajia Bai, Zhengxiao Xu, and Qingjie Zhu. Progress of electrical resistance tomography application in oil and gas reservoirs for development dynamic monitoring.Processes, 11(10):2950–, 2023

    work page 2023

  8. [8]

    Standardized low-resolution brain electromagnetic tomography (sLORETA): technical details.Methods and findings in experimental and clinical pharmacology, 24:5–12, 2002

    R D Pascual-Marqui. Standardized low-resolution brain electromagnetic tomography (sLORETA): technical details.Methods and findings in experimental and clinical pharmacology, 24:5–12, 2002

    work page 2002

  9. [9]

    On bias and its reduction via standardization in discretized electromagnetic source localization problems.Inverse Problems, 40(9):095002, July 2024

    Joonas Lahtinen. On bias and its reduction via standardization in discretized electromagnetic source localization problems.Inverse Problems, 40(9):095002, July 2024

    work page 2024

  10. [10]

    Standardized Kalman filtering for time serial source localization of simultaneous subcortical and cortical brain activity.www.techrxiv.org/722443/ UqPLPdTHCJ-oxYE2bCoI6Q, 2024

    Joonas Lahtinen, Paavo Ronni, Narayan Puthanmadam Subramaniyam, Alexandra Koulouri, Carsten Wolters, and Sampsa Pursiainen. Standardized Kalman filtering for time serial source localization of simultaneous subcortical and cortical brain activity.www.techrxiv.org/722443/ UqPLPdTHCJ-oxYE2bCoI6Q, 2024

    work page 2024

  11. [11]

    Evaluation of sLORETA in the presence of noise and multiple sources.Brain topography, 16(4):277–280, 2004

    Michael Wagner, Manfred Fuchs, and J¨ orn Kastner. Evaluation of sLORETA in the presence of noise and multiple sources.Brain topography, 16(4):277–280, 2004

    work page 2004

  12. [12]

    Analysis of the dynamics and origin of epileptic activity in patients with tuberous sclerosis evaluated for surgery of epilepsy.Clinical neurophysiology, 119(4):853–861, 2008

    Alberto J.R Leal, Ana I Dias, Jos´ e P Vieira, Ana Moreira, Lu´ ıs T´ avora, and Eul´ alia Calado. Analysis of the dynamics and origin of epileptic activity in patients with tuberous sclerosis evaluated for surgery of epilepsy.Clinical neurophysiology, 119(4):853–861, 2008

    work page 2008

  13. [13]

    sLORETA allows reliable distributed source reconstruction based on subdural strip and grid recordings.Human brain mapping, 33(5):1172–1188, 2012

    Matthias D¨ umpelmann, Tonio Ball, and Andreas Schulze-Bonhage. sLORETA allows reliable distributed source reconstruction based on subdural strip and grid recordings.Human brain mapping, 33(5):1172–1188, 2012

    work page 2012

  14. [14]

    de Gooijer-van de Groep, Frans S.S

    Karin L. de Gooijer-van de Groep, Frans S.S. Leijten, Cyrille H. Ferrier, and Geertjan J.M. Huiskamp. Inverse modeling in magnetic source imaging: Comparison of MUSIC, SAM(g2), and sLORETA to interictal intracranial EEG.Human brain mapping, 34(9):2032–2044, 2013. On Unbiased Source Parameter Estimation27

    work page 2032

  15. [15]

    Ana Coito, Silke Biethahn, Janina Tepperberg, Margherita Carboni, Ulrich Roelcke, Margitta Seeck, Pieter Mierlo, Markus Gschwind, and Serge Vulliemoz. Interictal epileptogenic zone localization in patients with focal epilepsy using electric source imaging and directed functional connectivity from low-density EEG.Epilepsia open, 4(2):281–292, 2019

    work page 2019

  16. [16]

    Vogrin, William P

    Rui Li, Chris Plummer, Simon J. Vogrin, William P. Woods, Levin Kuhlmann, Ray Boston, David T.J. Liley, Mark J. Cook, and David B. Grayden. Interictal spike localization for epilepsy surgery using magnetoencephalography beamforming.Clinical neurophysiology, 132(4):928–937, 2021

    work page 2021

  17. [17]

    Kwang Yeon Kim, Ja-Un Moon, Joo-Young Lee, Tae-Hoon Eom, Young-Hoon Kim, and In-Goo Lee. Distributed source localization of epileptiform discharges in juvenile myoclonic epilepsy: Standardized low-resolution brain electromagnetic tomography (sloreta) study.Medicine (Baltimore), 101(26):e29625–e29625, 2022

    work page 2022

  18. [18]

    Riedel, and Niels K

    Daniel van de Velden, Ev-Christin Heide, Caroline Bouter, Jan Bucerius, Christian H. Riedel, and Niels K. Focke. Effects of inverse methods and spike phases on interictal high-density EEG source reconstruction.Clinical neurophysiology, 156:4–13, 2023

    work page 2023

  19. [19]

    Candes, J

    E.J. Candes, J. Romberg, and T. Tao. Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information.IEEE transactions on information theory, 52(2):489–509, 2006

    work page 2006

  20. [20]

    Chartrand

    R. Chartrand. Nonconvex compressive sensing and reconstruction of gradient-sparse images: Random vs. tomographic fourier sampling. In2008 15th IEEE International Conference on Image Processing, pages 2624–2627. IEEE, 2008

    work page 2008

  21. [21]

    The split bregman method for l1-regularized problems.SIAM journal on imaging sciences, 2(2):323–343, 2009

    Tom Goldstein and Stanley Osher. The split bregman method for l1-regularized problems.SIAM journal on imaging sciences, 2(2):323–343, 2009

    work page 2009

  22. [22]

    Abascal, Lorena Cuss´ o, Juan Jos´ e Vaquero, and Manuel Desco

    Paula Montesinos, Juan Felipe P.J. Abascal, Lorena Cuss´ o, Juan Jos´ e Vaquero, and Manuel Desco. Application of the compressed sensing technique to self-gated cardiac cine sequences in small animals.Magnetic resonance in medicine, 72(2):369–380, 2014

    work page 2014

  23. [23]

    Interpreting magnetic fields on the brain: minimum norm estimates.Medical & biological engineering & computing, 32(1):35–42, 1994

    M.S H¨ am¨ al¨ ainen and R.J Ilmoniemi. Interpreting magnetic fields on the brain: minimum norm estimates.Medical & biological engineering & computing, 32(1):35–42, 1994

    work page 1994

  24. [24]

    Saturnino, Hartwig R

    Oula Puonti, Koen Van Leemput, Guilherme B. Saturnino, Hartwig R. Siebner, Kristoffer H. Madsen, and Axel Thielscher. Accurate and robust whole-head segmentation from magnetic resonance images for individualized head modeling.NeuroImage, 219:117044, 2020

    work page 2020

  25. [25]

    C. H. Wolters, L. Grasedyck, and W. Hackbusch. Efficient computation of lead field bases and influence matrix for the FEM-based EEG and MEG inverse problem.Inverse Problems, 20(4):1099–1116, 2004

    work page 2004

  26. [26]

    Ramon, P

    C. Ramon, P. H. Schimpf, and J. Haueisen. Influence of head models on EEG simulations and inverse source localizations.Biomed. Eng. Online, 5(10), 2006

    work page 2006

  27. [27]

    H¨ oltershinken, Tim Erdbr¨ ugger, and Carsten H

    Malte B. H¨ oltershinken, Tim Erdbr¨ ugger, and Carsten H. Wolters. sloreta is equivalent to single dipole scanning, 2024

    work page 2024

  28. [28]

    Effect of EEG electrode number on epileptic source localization in pediatric patients

    Abbas Sohrabpour, Yunfeng Lu, Pongkiat Kankirawatana, Jeffrey Blount, Hyunmi Kim, and Bin He. Effect of EEG electrode number on epileptic source localization in pediatric patients. Clinical neurophysiology, 126(3):472–480, 2015

    work page 2015

  29. [29]

    Effect of number and placement of EEG electrodes on measurement of neural tracking of speech.PloS one, 16(2):e0246769–, 2021

    Jair Montoya-Mart´ ınez, Jonas Vanthornhout, Alexander Bertrand, and Tom Francart. Effect of number and placement of EEG electrodes on measurement of neural tracking of speech.PloS one, 16(2):e0246769–, 2021

    work page 2021

  30. [30]

    Effects of a reduction of the number of electrodes in the EEG montage on the number of identified seizure patterns.Scientific reports, 12(1):4621–7, 2022

    Moritz Tacke, Katharina Janson, Katharina Vill, Florian Heinen, Lucia Gerstl, Karl Reiter, and Ingo Borggraefe. Effects of a reduction of the number of electrodes in the EEG montage on the number of identified seizure patterns.Scientific reports, 12(1):4621–7, 2022

    work page 2022

  31. [31]

    Andres Soler, Luis Alfredo Moctezuma, Eduardo Giraldo, and Marta Molinas. Automated methodology for optimal selection of minimum electrode subsets for accurate EEG source estimation based on genetic algorithm optimization.Scientific reports, 12(1):11221–18, 2022

    work page 2022

  32. [32]

    Horrillo-Maysonnial, T

    A. Horrillo-Maysonnial, T. Avigdor, C. Abdallah, D. Mansilla, J. Thomas, N. von Ellenrieder, On Unbiased Source Parameter Estimation28 J. Royer, B. Bernhardt, C. Grova, J. Gotman, and B. Frauscher. Targeted density electrode placement achieves high concordance with traditional high-density EEG for electrical source imaging in epilepsy.Clinical neurophysio...

    work page 2023

  33. [33]

    Standardized shrinking LORETA-FOCUSS (SSLOFO): a new algorithm for spatio-temporal EEG source reconstruction.IEEE transactions on biomedical engineering, 52(10):1681–1691, 2005

    Hesheng Liu, P.H Schimpf, Guoya Dong, Xiaorong Gao, Fusheng Yang, and Shangkai Gao. Standardized shrinking LORETA-FOCUSS (SSLOFO): a new algorithm for spatio-temporal EEG source reconstruction.IEEE transactions on biomedical engineering, 52(10):1681–1691, 2005

    work page 2005

  34. [34]

    Stable EEG source estimation for standardized Kalman filter using rate-of- change tracking.Computer methods and programs in biomedicine, 278:109280–, 2026

    Joonas Lahtinen. Stable EEG source estimation for standardized Kalman filter using rate-of- change tracking.Computer methods and programs in biomedicine, 278:109280–, 2026

    work page 2026

  35. [35]

    The effect of prior parameters on standardized Kalman filter-based EEG source localization.Biomedical signal processing and control, 121:110233–, 2026

    Dilshanie Prasikala, Joonas Lahtinen, Alexandra Koulouri, and Sampsa Pursiainen. The effect of prior parameters on standardized Kalman filter-based EEG source localization.Biomedical signal processing and control, 121:110233–, 2026

    work page 2026

  36. [36]

    Enhanced localization and orientation estimations in focal EEG source imaging using SVD-based coordinate transform.Brain Topography, 38(6):78, Oct 2025

    Joonas Lahtinen and Alexandra Koulouri. Enhanced localization and orientation estimations in focal EEG source imaging using SVD-based coordinate transform.Brain Topography, 38(6):78, Oct 2025

    work page 2025