Absement: Quantitative Assessment of Metabolic Cost during Quasi-Isometric Muscle Loading

Serhii V Marchenko

REVIEW 3 major objections 5 minor 19 references

Absement: Quantitative Assessment of Metabolic Cost during Quasi-Isometric Muscle Loading

T0 review · 3 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that the metabolic cost of quasi-isometric muscle loading is, to second order, a function of only two kinematic summaries of the muscle-length trajectory: the deviation absement and the integral of the squared deviation.

desk verdict Solid Taylor-expansion math for a quasi-static muscle model, but the paper overmarkets absement as a universal metabolic predictor; the testable moment-based protocol is the real contribution. read the letter →

arxiv 2512.13720 v2 pith:GDSIHVE4 submitted 2025-12-11 physics.bio-ph q-bio.QM

Serhii V Marchenko This is my paper · ORCID

classification physics.bio-phq-bio.QM

keywords absementmetaboliccostquasi-isometricmuscleloadingpostureholdingasymptoticexpansionactivationtremorindirectcalorimetry

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

During quasi-isometric posture holding, the paper claims, the total metabolic cost can be decomposed asymptotically into three terms: a baseline cost P0T, a first-order cost proportional to the deviation absement (the time integral of muscle-length deviation, which equals T times the mean deviation), and a second-order cost proportional to the integral of the squared deviation (T times the sum of squared mean and variance). The result follows from eliminating muscle activation through joint-moment equilibrium, reducing the cost functional to an integral of a single smooth length-dependent function, and Taylor-expanding it. If correct, this means that no trajectory detail beyond the mean and variance of length deviation matters to second order, and that the parameters can be recovered by linear regression from ordinary kinematic recordings and indirect calorimetry. This matters because traditional scalar metrics such as time under tension ignore the temporal structure that the paper shows is energetically significant.

What carries the argument

The carrying device is the reduction of the cost functional to a scalar function of length. The quasi-static moment balance Q(θ,a)=0 is solved locally for activation a*(θ) by the implicit-function theorem (with non-degeneracy condition Q_a≠0), so metabolic power becomes φ(ℓ) = α a*(Θ(ℓ)) + β F(ℓ, a*(Θ(ℓ))) after inversion to length. Taylor expansion of ∫φ(ℓ(t))dt to second order yields the theorem; the first variation is necessarily an integral against a constant kernel, which is what forces the absement (the time integral of displacement, a named quantity from integral kinematics) to be the unique linear descriptor, and the second-order term to be the integral of the squared deviation.

What would settle it

Compare metabolic cost (indirect calorimetry) for two protocols that match the mean and variance of muscle-fascicle length deviation but differ in the frequency or ordering of deviations (e.g., slow drift vs. high-frequency tremor with same amplitude distribution). If the measured energy differs beyond experimental error, the second-order truncation is wrong. A more direct test: vary activation at fixed length and vary force at fixed activation separately to check whether the joint cost is linear in (a, F), as the model requires.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1: under the model assumptions, the metabolic energy functional E_met(ℓ) admits, for small deviations from a reference posture, the asymptotic representation E_met(ℓ) = P0 T + C1 ΔA_ℓ + C2 ∫_0^T (ℓ(t)−ℓ0)^2 dt + O(‖ℓ−ℓ0‖^3_{L∞}), where ΔA_ℓ = ∫_0^T (ℓ(t)−ℓ0) dt is the deviation absement and the constants are fixed by local derivatives of the force law, moment arm, and external moment at equilibrium. Lemma 1 sharpens the first-order term: any linear functional that matches the first variation of E_met must be proportional to ΔA_ℓ, making absement the unique first-order sufficient statistic. The paper presents this as a fundamental, not phenomenological, variable:

Load-bearing premise

The load-bearing premise is that metabolic power is exactly a linear combination of activation and force, P_met = α a + β F with constant positive coefficients; if real muscle energetics are nonlinear in activation or force (or history-dependent), the reduction of cost to a function of length alone fails.

Editorial extensions

If this is right

The optimal posture-holding strategy to first order is to keep the time-averaged muscle length at the reference value; once mean drift is minimized, the remaining cost is reduced by minimizing length variance (tremor amplitude).
A practical parameter-identification protocol follows directly: record muscle length (e.g., ultrasound) or joint angle, compute T, ΔA_ℓ, and ∫x²dt, and fit multiple linear regression against measured metabolic cost to recover P0, C1, C2.
No separate cycle-specific predictor is needed for periodic variations within the model: periodicity contributes only through its mean (absement) and dispersion (second moment).
The three-term decomposition gives a physical interpretation: P0T is the cost of ideal holding, C1ΔA_ℓ the cost of systematic drift, and C2∫x²dt the cost of tremor/variability.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A direct experimental falsification suggested by the paper's structure: two protocols with identical mean and variance of length deviation but different frequency content should have identical metabolic cost under the theorem; if calorimetry shows a difference, the quasi-static reduction breaks at the second-order level.
In a multi-joint extension, the paper's proof structure suggests a vector-valued absement is the first-order coordinate, one per joint, with cross-coupling in the quadratic term.
The linear-cost premise could be probed by independent measurement of activation-dependent and force-dependent cost; the theorem requires the coefficients α and β to be constant across the trajectory.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The math here is correct and clearly presented. The reduction of the cost functional to ∫φ(ℓ)dt under quasi-static equilibrium, the implicit function theorem step, and the Taylor expansion with the cubic remainder all check out. Lemma 1 and Theorem 1 are proven cleanly in the appendices, and the paper is honest about several limitations, especially in §5.2–5.3. The genuinely new and useful piece is the moment-based regression protocol: predict metabolic cost from T, ΔAℓ, and ∫δℓ²dt, then test for residual frequency dependence. That is a crisp, falsifiable hypothesis and worth taking seriously.

The soft spot is not the derivation—it is the claim that absement is the "unique first-order sufficient statistic" for real quasi-isometric loading. The stress-test note gets this right. What the proof actually requires is that metabolic power be an instantaneous function of the state (ℓ, a), with no dependence on ℓ̇ or history. The linear-in-activation-and-force form is not the essential assumption; any smooth instantaneous function reduces to φ(ℓ), so the Taylor structure survives. But real muscle energetics has velocity-dependent and history-dependent terms: Fenn effect, shortening/lengthening heat, force depression/enhancement, activation dynamics. As the stress-test note shows, even a minimal velocity-dependent term P_met = φ(ℓ) + h(ℓ)ℓ̇ + γℓ̇² changes the first-order term to a boundary contribution h(ℓ0)[ℓ(T)−ℓ(0)] and adds γ∫ℓ̇²dt at second order. Two trajectories can share the same mean and variance yet differ in ∫ℓ̇²dt, so the theorem's conclusion that only mean and variance matter does not survive for real muscle. The paper itself says in §5.3 that experimental validation is future work and that an extended model should include velocity-dependent terms, so the limitation is acknowledged—but the abstract and conclusions still market absement as universal. That overreach should be fixed.

The regression coefficients are fitted, not predicted, so "quantitative assessment" is actually "a plausible model with fitted coefficients." That is fine if stated that way. The proposed residual-frequency test is exactly the right way to probe the quasi-static assumption; if frequency dependence vanishes after controlling for mean and variance, the model wins. If not, the model has a boundary.

My take: this deserves a serious referee, but a conditional accept with requests to temper the uniqueness claims and to frame the result as a quasi-static approximation with a built-in falsification test. I would bring it to a reading group for the discussion of when Taylor expansions become "sufficient statistics" and when they do not.

Referee Report

3 major / 5 minor

Summary. The paper introduces deviation absement ΔA_ℓ = ∫₀ᵀ(ℓ(t)−ℓ₀)dt as a first-order predictor of metabolic cost during quasi-isometric muscle loading. Under a single-degree-of-freedom quasi-static model with metabolic power P_met = αa + βF(ℓ,a), the paper eliminates activation via the implicit function theorem, reduces the energy functional to ∫φ(ℓ)dt, and proves an asymptotic expansion E_met = P₀T + C₁ΔA_ℓ + C₂∫₀ᵀδℓ²dt + O(‖δℓ‖³_L∞). It then proposes a linear-regression identification scheme for P₀, C₁, C₂ and argues that a residual dependence on oscillation frequency would mark the limits of the quasi-static reduction.

Significance. The mathematical derivation is elementary but correct under the stated assumptions: the reduction to a one-dimensional integral and the Taylor expansion are cleanly presented, and the proof in Appendix B is explicit. The main strength is a concrete, falsifiable consequence: within the model, only the mean and variance (equivalently absement and the second raw moment) of length deviation enter at second order, so a frequency-dependent residual would identify the failure of the quasi-static premise. However, the claimed 'unique first-order sufficient statistic' is conditional on the instantaneous-state cost model of Eq. (3); the paper itself acknowledges (§5.3) that experimental validation is future work. As a theoretical contribution with a testable prediction the paper is valuable, but the abstract and conclusions overstate its status as a quantitative assessment of real muscle energetics.

major comments (3)

[§2.3, Eq. (3); Theorem 1] The uniqueness claim is an artifact of the instantaneous-state cost model. The reduction to E_met(ℓ)=∫φ(ℓ)dt requires P_met to be a function of state (ℓ,a) only. If a velocity-dependent cost is admitted, e.g. P_met=φ(ℓ)+h(ℓ)ℓ̇+γℓ̇², the first-order term becomes h(ℓ₀)[ℓ(T)−ℓ(0)] rather than ΔA_ℓ, and the second-order term gains γ∫ℓ̇²dt. Two trajectories can have identical ΔA_ℓ and ∫δℓ²dt but different ∫ℓ̇²dt, so absement is no longer the unique first-order kinematic statistic. Since real muscle energetics includes Fenn/heat and history-dependent terms, the paper's abstract and conclusion should explicitly frame the result as conditional on the instantaneous-state assumption, and should not claim uniqueness for real muscle without further defense. The proposed frequency-residual test is a good probe, but the manuscript should not market absement as a universal sufficient statistic in advan
[§3.3, C₂ formula] The explicit formula C₂ = β/(2r₀²)(Fθθ + 2FθaCθ + FaaCθ²) is incomplete. In Appendix B the full second-order coefficient includes contributions from the curvature of a*(θ), namely αB + βFaB with B = ½a*''(θ₀), and also the ℓθθ term in Fℓκ₀. The §3.3 derivation approximates a*(θ) only to first order before computing the quadratic term, which is inconsistent with a genuine second-order expansion. Because C₂ is one of the three regression coefficients in the proposed identification scheme, this discrepancy is not a cosmetic detail; the formula should be corrected, or §3.3 should state explicitly that the full coefficient is given in Appendix B.
[Abstract; §5.3] The paper describes the result as enabling 'quantitative assessment of metabolic cost' and claims the formalism makes it possible to 'recover physically meaningful coefficients ... by means of linear regression,' yet §5.3 lists experimental validation as future work. Without validation or at least a synthetic-data demonstration of the identification procedure, these claims overreach. The manuscript should either present the result as a conditional theoretical hypothesis with a clearly specified falsification protocol, or include a numerical check of the regression scheme on simulated trajectories to support the practical claims.

minor comments (5)

[§A] The proof of Lemma 1 states that any linear integral functional on C[0,T] can be represented as ∫k(t)h(t)dt. This is not true for arbitrary continuous linear functionals (which are represented by signed measures); it holds only for integral functionals with a bounded measurable kernel. Since the lemma concerns integral descriptors, the wording should be tightened to avoid an invalid generalization.
[Throughout] The phrase 'sufficient statistic' is used in a non-statistical sense. Consider 'unique linear integral descriptor' or 'leading asymptotic coordinate' to avoid confusion with the statistical term.
[§2.3] The parameters α, β, and the reference length ℓ₀ are free parameters, so the 'universal' expansion is universal only within a model family with three undetermined coefficients. This should be stated explicitly in the abstract.
[References] References [14] and [19] are from hydraulophone music and earthquake early warning; their connection to muscle energetics is loose. The authors may wish to clarify the analogy or shorten that discussion.
[§3.3] The notation F is used both for the force function F(ℓ,a) and for the composed function F(θ,a)=F(ℓ(θ),a). This is convenient but can confuse the reader when derivatives such as Fθθ are introduced; a short remark or distinct symbol would help.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the expansion is a direct Taylor consequence of explicitly stated modeling assumptions, and the fitted parameters are not presented as independent predictions.

full rationale

The paper's central derivation is self-contained analytical work: given Eq. (3), P_met = αa + βF(ℓ,a), the quasi-static equilibrium Q=0 with Q_a≠0 yields a=a*(θ), and with r0≠0 yields P_met=φ(ℓ). Theorem 1 is then the standard Taylor expansion of ∫φ(ℓ)dt; the appearance of absement ∫δℓ dt and the quadratic moment ∫δℓ²dt is a mathematical consequence, not an input. The 'uniqueness' in Lemma 1 is a direct statement about first variations of that integral functional and does not invoke any unverified self-citation or imported uniqueness theorem. The proposed regression in §4.2 estimates P0, C1, C2 from the same data, but the paper labels this as parameter identification rather than an independent prediction, and it explicitly marks velocity/history dependence and experimental validation as limitations/future work (§5.2, §5.3). No load-bearing self-citations exist. The linear-cost ansatz is a model assumption, not a circular step; the paper's honesty about its scope prevents the derivation from being a disguised fit.

Assumptions & free parameters 3 free parameters · 5 assumptions · 1 invented entities

The central expansion rests on three classes of input: standard calculus assumptions (smoothness), domain assumptions (quasi-static, small deviations, non-degenerate activation), and one ad hoc modeling assumption (linear metabolic power in activation and force). The free parameters α, β, and the reference length ℓ0 are not derived from first principles; they must be fitted to calorimetric data in the proposed protocol. No new physical entity is introduced beyond a renamed statistic.

free parameters (3)

α = unspecified
Cost per unit activation in Eq. (3); positive constant, must be identified from regression of metabolic data.
β = unspecified
Cost per unit force in Eq. (3); positive constant, must be identified from regression of metabolic data.
ℓ0 (reference length) = unspecified
The equilibrium length about which deviations are measured; its choice affects ΔAℓ and the fitted coefficients.

assumptions (5)

standard math F, r, M_ext, ℓ(θ) are smooth (C² or C³) functions in a neighbourhood of the equilibrium point
Invoked in §2.2, §3.1, and Appendix B to justify Taylor expansions and the implicit function theorem.
domain assumption Q_a(θ0,a0) = F_a r0 ≠ 0 (activation can change joint moment at fixed angle)
Eq. (2); required for the implicit function theorem to eliminate activation. Excludes passive or saturated muscle states.
domain assumption Quasi-static (quasi-isometric) regime: inertial and viscous effects neglected
Section 2.2; permits Q(θ,a)=0 at each instant and makes power a function of instantaneous state.
ad hoc to paper Metabolic power is an instantaneous linear function of activation and force: P_met = αa + βF
§2.3 Eq. (3); not derived from physiology and no citation provided. This is the load-bearing premise for the expansion.
domain assumption Small deviations: ‖ℓ−ℓ0‖_{L∞} ≤ ε with ε → 0
§3.1; required for uniform remainder estimates in the expansion.

invented entities (1)

deviation absement ΔAℓ
purpose: Named first-order kinematic predictor of metabolic cost; time integral of length deviation.
A renamed/repurposed quantity from engineering (absement); here it equals T times the mean deviation, so it is not an independent physical entity but a bookkeeping variable.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Absement: Quantitative Assessment of Metabolic Cost during Quasi-Isometric Muscle Loading." pith.science (2026). https://pith.science/paper/GDSIHVE4

@misc{pith2026251213720,
  author       = {Pith},
  title        = {Pith review of: Absement: Quantitative Assessment of Metabolic Cost during Quasi-Isometric Muscle Loading},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GDSIHVE4}},
  note         = {Machine review of arXiv:2512.13720}
}

abstract

Small deviations during nominally isometric loading can be separated into a sustained mean offset and fluctuations about that offset. We develop a local quasi-static model that connects these video-accessible kinematic quantities to metabolic energy. Muscle activation is eliminated through joint-moment equilibrium, and the resulting metabolic power is reduced to a smooth function of a locally invertible muscle-length coordinate. For the deviation \(x(t)=\ell(t)-\ell_0\), the reduced energy satisfies \[ \mathcal{E}_{\mathrm{met}}[\ell] = P_0T + C_1\Delta\mathcal{A}_{\ell} + C_2\mathcal{M}_{2,\ell} + R_3, \qquad |R_3| \leq KT\|x\|_{L^\infty(0,T)}^3, \] where \(\Delta\mathcal{A}_{\ell}=\int_0^T x(t)\,dt\) is signed deviation absement and \(\mathcal{M}_{2,\ell}=\int_0^T x(t)^2\,dt\) is the second raw integral moment. Equivalently, if \(\mu_\ell\) and \(\sigma_\ell^2\) are the mean and variance of the observed length deviation, then \(\Delta\mathcal{A}_{\ell}=T\mu_\ell\) and \(\mathcal{M}_{2,\ell}=T(\mu_\ell^2+\sigma_\ell^2)\). A video-based protocol can therefore estimate the required predictors without differentiating the recorded trajectory. Within the autonomous quasi-static model, periodic variation does not require a separate cycle-specific predictor: its mean contributes through absement and its dispersion through the second moment. This moment-based protocol, rather than the Taylor expansion alone, provides the experimentally testable result: residual dependence on frequency after control for the first two moments would identify the limit of the quasi-static reduction.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 1 linked inside Pith

[1]

Lagrangian for circuits with higher-order elements.Entropy, 21(12):1230, 2019

Dalibor Biolek, Zdenˇ ek Biolek, and Viera Biolkov´ a. Lagrangian for circuits with higher-order elements.Entropy, 21(12):1230, 2019

2019
[2]

de Hijas-Liste, Edda Klipp, Eva Balsa-Canto, and Julio R

Gundi´ an M. de Hijas-Liste, Edda Klipp, Eva Balsa-Canto, and Julio R. Banga. Global dynamic optimization approach to predict activation in metabolic pathways.BMC Systems Biology, 8(1):1, 2014

2014
[3]

Delp, Frank C

Scott L. Delp, Frank C. Anderson, Allison S. Arnold, J. Peter Loan, Ashraf Habib, Chand T. John, Emma Guendelman, and Darryl G. Thelen. Opensim: Open-source software to create and analyze dynamic simulations of movement.IEEE Transactions on Biomedical Engineering, 54(11):1940–1950, 2007

1940
[4]

The modelling of metabolic systems: Structure, control and optimal design.BioSystems, 47(1-2):61–77, 1998

Reinhart Heinrich and Stefan Schuster. The modelling of metabolic systems: Structure, control and optimal design.BioSystems, 47(1-2):61–77, 1998

1998
[5]

Brown, and Jaap H

Han Houdijk, Starr E. Brown, and Jaap H. van Die¨ en. Relation between postural sway mag- nitude and metabolic energy cost during upright standing on a compliant surface.Journal of Applied Physiology, 119(6):696–703, 2015

2015
[6]

Trienke Ijmker, Han Houdijk, Claudine J. C. Lamoth, Peter J. Beek, and Lucas H. V. van der Woude. Energy cost of balance control during walking decreases with external stabilizer stiff- ness independent of walking speed.Journal of Biomechanics, 46(13):2109–2114, 2013

2013
[7]

Memory elements: A paradigm shift in lagrangian modeling of electrical circuits.IF AC Proceedings Volumes, 45(2):25–30, 2012

Dimitri Jeltsema. Memory elements: A paradigm shift in lagrangian modeling of electrical circuits.IF AC Proceedings Volumes, 45(2):25–30, 2012

2012
[8]

Matthew C. Kelley. Acoustic absement in detail: Quantifying acoustic differences across time- series representations of speech data. arXiv preprint arXiv:2304.06183, 2023

arXiv 2023

Show all 19 references

[9]

Kelley and Benjamin V

Matthew C. Kelley and Benjamin V. Tucker. Using acoustic distance and acoustic absement to quantify lexical competition.The Journal of the Acoustical Society of America, 151(2):1367– 1379, 2022. 20

2022
[10]

Competition for enzymes in metabolic pathways: Implica- tions for optimal distributions of enzyme concentrations and for the distribution of flux control

Edda Klipp and Reinhart Heinrich. Competition for enzymes in metabolic pathways: Implica- tions for optimal distributions of enzyme concentrations and for the distribution of flux control. BioSystems, 54(1-2):1–14, 1999

1999
[11]

Pinto, Joanna Diong, and Robert D

Li Khim Kwah, Rafael Z. Pinto, Joanna Diong, and Robert D. Herbert. Reliability and validity of ultrasound measurements of muscle fascicle length and pennation in humans: A systematic review.Journal of Applied Physiology, 114(6):761–769, 2013

2013
[12]

Steve Mann, M. L. Hao, M. Tsai, M. Hafezi, A. Azad, and F. Keramatimoezabad. Effectiveness of integral kinesiology feedback for fitness-based games. In2018 IEEE Games, Entertainment, Media Conference (GEM), pages 1–9. IEEE, 2018

2018
[13]

Integral kinematics (time-integrals of distance, energy, etc.) and integral kinesiology

Steve Mann and Ryan Janzen. Integral kinematics (time-integrals of distance, energy, etc.) and integral kinesiology. In2014 IEEE Games, Entertainment, Media Conference (GEM), pages 1–8. IEEE, 2014

2014
[14]

Hydraulophone design considerations: Absement, displacement, and velocity-sensitive music keyboard in which each key is a water jet

Steve Mann, Ryan Janzen, and Mark Post. Hydraulophone design considerations: Absement, displacement, and velocity-sensitive music keyboard in which each key is a water jet. In Proceedings of the 14th ACM International Conference on Multimedia, pages 519–528. ACM, 2006

2006
[15]

Miles-Chan and Abdul G

Jennifer L. Miles-Chan and Abdul G. Dulloo. Posture allocation revisited: Breaking the sedentary threshold of energy expenditure for obesity management.Frontiers in Physiology, Volume 8 - 2017, 2017

2017
[16]

Monnard and Jennifer L

Cathr ´ ıona R. Monnard and Jennifer L. Miles-Chan. Energy cost of standing in a multi-ethnic cohort: Are energy-savers a minority or the majority?PLOS ONE, 12(1):1–12, 01 2017

2017
[17]

Floh, and Christopher S

Hala Mtaweh, Sanna Tuira, Armin A. Floh, and Christopher S. Parshuram. Indirect calorime- try: History, technology, and application.Frontiers in Pediatrics, 6:257, 2018

2018
[18]

Rosa, Jonathan S

Luis G. Rosa, Jonathan S. Zia, Omer T. Inan, and Gregory S. Sawicki. Machine learning to extract muscle fascicle length changes from dynamic ultrasound images in real-time.PLOS ONE, 16(5):1–17, 05 2021

2021
[19]

Magnitude determination using cumulative absolute absement for earthquake early warning.Geoscience Letters, 10(1):1, 2023

Yih-Min Wu, Himanshu Mittal, Yueh-Ho Lin, and Yu-Hsuan Chang. Magnitude determination using cumulative absolute absement for earthquake early warning.Geoscience Letters, 10(1):1, 2023. 21

2023

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [1]

Lagrangian for circuits with higher-order elements.Entropy, 21(12):1230, 2019

Dalibor Biolek, Zdenˇ ek Biolek, and Viera Biolkov´ a. Lagrangian for circuits with higher-order elements.Entropy, 21(12):1230, 2019

2019

[2] [2]

de Hijas-Liste, Edda Klipp, Eva Balsa-Canto, and Julio R

Gundi´ an M. de Hijas-Liste, Edda Klipp, Eva Balsa-Canto, and Julio R. Banga. Global dynamic optimization approach to predict activation in metabolic pathways.BMC Systems Biology, 8(1):1, 2014

2014

[3] [3]

Delp, Frank C

Scott L. Delp, Frank C. Anderson, Allison S. Arnold, J. Peter Loan, Ashraf Habib, Chand T. John, Emma Guendelman, and Darryl G. Thelen. Opensim: Open-source software to create and analyze dynamic simulations of movement.IEEE Transactions on Biomedical Engineering, 54(11):1940–1950, 2007

1940

[4] [4]

The modelling of metabolic systems: Structure, control and optimal design.BioSystems, 47(1-2):61–77, 1998

Reinhart Heinrich and Stefan Schuster. The modelling of metabolic systems: Structure, control and optimal design.BioSystems, 47(1-2):61–77, 1998

1998

[5] [5]

Brown, and Jaap H

Han Houdijk, Starr E. Brown, and Jaap H. van Die¨ en. Relation between postural sway mag- nitude and metabolic energy cost during upright standing on a compliant surface.Journal of Applied Physiology, 119(6):696–703, 2015

2015

[6] [6]

Trienke Ijmker, Han Houdijk, Claudine J. C. Lamoth, Peter J. Beek, and Lucas H. V. van der Woude. Energy cost of balance control during walking decreases with external stabilizer stiff- ness independent of walking speed.Journal of Biomechanics, 46(13):2109–2114, 2013

2013

[7] [7]

Memory elements: A paradigm shift in lagrangian modeling of electrical circuits.IF AC Proceedings Volumes, 45(2):25–30, 2012

Dimitri Jeltsema. Memory elements: A paradigm shift in lagrangian modeling of electrical circuits.IF AC Proceedings Volumes, 45(2):25–30, 2012

2012

[8] [8]

Matthew C. Kelley. Acoustic absement in detail: Quantifying acoustic differences across time- series representations of speech data. arXiv preprint arXiv:2304.06183, 2023

arXiv 2023