Skip to main content

m13 · Catheter design: electrode size and spacing, high-density grids, baskets, contact-force sensing

Warning

Education only. Not for diagnosis, treatment or any clinical decision.

In 60 seconds

A mapping catheter is a sampling instrument. Its electrodes’ size sets how much they average and how much spreading resistance they present to the blood: a flush disk of radius \(a\) has \(1/(4\sigma a)\), so halving the size doubles it. Its spacing and its height above the activated tissue set what it can resolve: the sharpest signal a front can make, seen from height \(h\), has spectrum \(e^{-|k|h}\), and electrodes at spacing \(\Delta\) lose the fraction \(e^{-2\pi h/\Delta}\) of its energy above the spatial Nyquist limit. Bipoles see only the gradient along their axis, so a high-density grid that forms perpendicular pairs can remove most of the direction dependence. A basket spreads electrodes over a chamber, with its splines farthest apart at the equator and further apart still if they bunch. A contact-force sensor reads a few strains around a small spring element and solves a least-squares problem for the force; the sensor layout decides how noise turns into force error.

From the field

  • Electrode size changes the map. In 20 patients with scar-related atrial arrhythmias, mapping with 1-mm electrodes found a smaller area of bipolar voltage below 0.5 mV than mapping with a 3.5-mm tip (14.7 versus 20.4 cm²), and distinct, annotatable electrograms at 54.4% versus 21.4% of low-voltage points (Anter et al. 2015).
  • Height above the tissue matters most. A modeling study found that electrode diameter, length, spacing and height all degraded spatial resolution, height the most; electrodes of negligible size would improve resolution by about one order of magnitude at most (Stinnett-Donnelly et al. 2012).
  • Baskets rarely open evenly. In 25 patients, deployed baskets showed equatorial bunching, with interspline spacing from 1.5 to 85 mm in the left atrium, and about one third of electrodes failed to show a median peak-to-peak voltage above twice the low-voltage threshold (Laughner et al. 2016).
  • Contact force is measured, not guessed. One catheter design uses three optical fibers to measure microdeformation of the tip; on the bench its readings tracked a certified balance from 0 to 50 g with a mean error of 1.0 g or less (Yokoyama et al. 2008).

Derivations

Symbols used throughout. \(a\) electrode radius (mm); \(\sigma\) blood conductivity; \(R_s\) spreading resistance; \(\theta\) angle between a bipole’s axis and the front’s direction of travel; \(A_0\) the amplitude of a bipole aligned with it; \(h\) height of the electrode above the activated layer (mm); \(x\) distance across a front (mm); \(k\) spatial wavenumber (rad/mm); \(\Delta\) electrode spacing (mm); \(E\) fraction of signal energy above the spatial Nyquist limit; \(R\) basket radius (mm); \(n\) number of splines; \(\vartheta\) polar angle on the basket; \(\varphi_i\) angle of strain sensor \(i\); \(F_x\), \(F_y\) lateral and \(F_z\) axial force; \(\alpha\), \(\beta\) axial and lateral sensor sensitivities; \(s_i\) sensor reading; \(\sigma_s\) sensor noise SD.

Electrode size and spreading resistance

Current leaving a small electrode crowds near it, so much of the resistance to a distant return lies within a few radii: for a sphere, two thirds of it within three radii. For an equipotential disk flush with an insulating plane, the current density is proportional to \(1/\sqrt{a^2 - r^2}\), and Newman showed the resistance is (Newman 1966):

\[ R_s^{\text{disk}} = \frac{1}{4\sigma a}, \qquad R_s^{\text{hemisphere}} = \frac{1}{2\pi\sigma a}, \qquad R_s^{\text{sphere}} = \frac{1}{4\pi\sigma a} \tag{1}\]

The sphere is the infinite-medium tip model of m27 (eq-m27-spreading). All three scale as \(1/a\): halve the electrode and the resistance doubles. That resistance makes thermal noise, which grows as its square root (m14). The lab checks numerically that the \(1/\sqrt{a^2 - r^2}\) density makes the disk an equipotential and reproduces \(1/(4\sigma a)\). Real electrodes are rings and capped cylinders on a catheter body, and at low frequencies the electrode–electrolyte interface adds its own impedance; these formulas give the scaling, not a specification.

Key takeaways

  • Spreading resistance is set near the electrode and scales as \(1/(\text{radius})\).
  • Smaller electrodes average less (m03) but present more resistance.

Bipole direction and high-density grids

A closely spaced bipole records the potential difference along its axis, which for a planar front is proportional to the gradient component along that axis (m03, eq-m03-bipolar-planar):

\[ A(\theta) = A_0\,\lvert\cos\theta\rvert, \qquad \max\bigl(\lvert\cos\theta\rvert, \lvert\sin\theta\rvert\bigr) \ge \tfrac{1}{\sqrt2}, \qquad \sqrt{A(\theta)^2 + A(\theta - \tfrac{\pi}{2})^2} = A_0 \tag{2}\]

A single bipole can read zero when it lies along the front. A grid that forms two perpendicular bipoles at one site never loses more than 29% with the better of the two, and their combination is independent of direction. Omnipolar methods take this further by using the whole electric-field loop (m03, eq-m03-omnipole) (Deno et al. 2017; Massé et al. 2016). For curved fronts, unequal spacings or noise, the identity holds only approximately.

Key takeaways

  • One bipole’s amplitude depends on direction; two perpendicular bipoles largely remove the dependence.

Spacing, height and spatial aliasing

A planar front is a line of dipoles. Across its width, the gradient signal it makes at height \(h\) above the activated layer is the Poisson kernel for the half-plane, the sharpest pattern Laplace’s equation allows at that height. Its spectrum decays exponentially (m03, eq-m03-height-filter), and electrodes at spacing \(\Delta\) capture wavenumbers only up to \(\pi/\Delta\):

\[ g(x) = \frac{h/\pi}{x^2 + h^2}, \quad \hat g(k) = e^{-\lvert k\rvert h}, \qquad E = \frac{\int_{\pi/\Delta}^{\infty} e^{-2kh}\,dk}{\int_0^{\infty} e^{-2kh}\,dk} = e^{-2\pi h/\Delta}, \qquad \Delta_{\max} = \frac{2\pi h}{\ln(1/E)} \tag{3}\]

The energy fraction \(E\) above the Nyquist limit is lost or folded back as aliasing. Spacing should scale with height: at \(h\) = 1 mm, keeping \(E\) below 1% needs \(\Delta \le\) 1.36 mm, and at \(h\) = 2 mm, 2.73 mm (lab part C). These heights are illustrative; \(h\) includes tissue depth and any gap between electrode and tissue.

Exact

Assumptions

  • Model: the half-plane Poisson kernel and its spectrum, Equation 3, sampled on a uniform line of point electrodes. Tested against epml.catheters.
  • Parameters: spacing, height and grid offset are free choices for illustration.
  • What it ignores: electrode size (m03), the unipolar far field, curved fronts, tissue anisotropy and noise.

Key takeaways

  • Height sets the signal’s bandwidth; spacing must follow it.
  • The fraction above Nyquist falls exponentially with \(h/\Delta\).

Basket geometry

A basket’s splines run from pole to pole. With \(n\) splines evenly spread around the axis, the arc between neighbors at polar angle \(\vartheta\) is:

\[ s(\vartheta) = \frac{2\pi R\sin\vartheta}{n} \tag{4}\]

It is largest at the equator: 19.6 mm for 8 splines on an illustrative 25-mm radius. Coverage is better judged by the fill distance, the largest distance from any wall point to its nearest electrode. For an 8 × 8 basket with electrodes from 25° to 155° polar angle, the fill distance over that band is 10.4 mm when the splines are even and 39.0 mm when they bunch into half the circumference (lab part D). Real atria are not spheres, and electrodes that do not touch the wall record far-field signals (Laughner et al. 2016).

Key takeaways

  • Spline spacing grows toward the equator and with poor deployment.
  • Fill distance, not electrode count, measures coverage.

Contact-force sensing

A contact-force tip has a small spring element and a few sensors that read its deformation. Model three sensors at angles \(\varphi_i\), each reading the axial force plus the lateral force projected on its direction:

\[ s_i = \alpha F_z + \beta\,(F_x\cos\varphi_i + F_y\sin\varphi_i) + \text{noise}, \qquad \mathbf s = C\mathbf F, \qquad \hat{\mathbf F} = (C^\top C)^{-1}C^\top\mathbf s \tag{5}\]

With independent noise of SD \(\sigma_s\), the estimate’s covariance and, for three sensors 120° apart, its standard deviations are:

\[ \operatorname{cov}(\hat{\mathbf F}) = \sigma_s^2\,(C^\top C)^{-1}, \qquad \operatorname{SD}(\hat F_x) = \operatorname{SD}(\hat F_y) = \sqrt{\tfrac{2}{3}}\,\frac{\sigma_s}{\beta}, \qquad \operatorname{SD}(\hat F_z) = \frac{\sigma_s}{\sqrt3\,\alpha} \tag{6}\]

Uneven layouts are worse: with sensors at 0° and ±60°, the error in \(F_x\) triples (lab part E), and sensors on one line cannot sense force perpendicular to it. Real sensors are calibrated, not modeled, and drift with temperature; the calibration matrix plays the role of \(C\).

Exact

Assumptions

  • Model: three linear sensors, Equation 5, with independent Gaussian noise of one unit; covariance by Equation 6. Tested against epml.catheters.
  • Parameters: sensor angles and the axial-to-lateral sensitivity ratio are free choices; lateral sensitivity is 1.
  • What it ignores: calibration error, temperature drift, nonlinearity, hysteresis and catheter dynamics.

Key takeaways

  • Force comes from a least-squares fit; the sensor layout sets its noise.
  • Symmetric layouts minimize the condition number.

Clinical correlate

NotePending clinical review

This section describes published findings only. It is not guidance for patient care, and it has not yet been reviewed by an EP physician.

Where this shows up in EP. Catheter choice shapes every map. Smaller, closer electrodes recorded more distinct electrograms and less apparent low voltage in scar-related atrial arrhythmias (Anter et al. 2015); deployed baskets often left large gaps and many low-voltage electrodes (Laughner et al. 2016). Contact force was a major determinant of lesion size in an animal preparation (Yokoyama et al. 2008). In a 32-patient study of pulmonary vein isolation, all 5 patients treated with an average force below 10 g had recurrences, while 8 of 10 treated above 20 g were free from recurrence at 12 months (Reddy et al. 2012). That is a small observational association, not proof that a force target causes better outcomes.

What the evidence is. A retrospective clinical comparison (Anter et al. 2015), a modeling study (Stinnett-Donnelly et al. 2012), a clinical deployment analysis (Laughner et al. 2016), a bench and animal study (Yokoyama et al. 2008) and a small multicenter clinical study (Reddy et al. 2012).

What a reader should ask:

  • What were the electrode sizes, spacings and heights? Resolution depends on all three (Equation 3).
  • Was voltage measured with bipoles in one direction only? (Equation 2)
  • How well did the basket deploy? Fill distance, not electrode count (Equation 4).
  • How was the force sensor calibrated, and what is its error? (Equation 6)

Visual summary

Schematic

smaller electrode:higher spreading resistance two perpendicular bipoles splines farthest apartat the equator three sensors solvefor the force
Figure 1: Four design choices: electrode size, bipole direction, basket spacing and force sensing.

Assumptions

How experts think

Before trusting a map or a catheter specification, ask:

  1. How high above the activated tissue are the electrodes? It sets the bandwidth (Equation 3).
  2. Is the spacing matched to that height? (Equation 3)
  3. Which way do the bipoles point relative to the front? (Equation 2)
  4. Where are the gaps? Fill distance, especially for baskets (Equation 4).
  5. What is the force sensor’s error, and how was it calibrated? (Equation 6)

One sentence to keep: a catheter is a spatial sampler; height, spacing and direction decide what it can see.

Cheat sheet

Quantity Result
Spreading resistance (Equation 1) disk \(1/(4\sigma a)\); hemisphere \(1/(2\pi\sigma a)\); sphere \(1/(4\pi\sigma a)\)
Bipole direction (Equation 2) \(A_0\lvert\cos\theta\rvert\); better of two perpendicular \(\ge A_0/\sqrt2\); RSS \(= A_0\)
Energy above Nyquist (Equation 3) \(E = e^{-2\pi h/\Delta}\); \(\Delta_{\max} = 2\pi h/\ln(1/E)\)
Basket spacing (Equation 4) \(2\pi R\sin\vartheta/n\)
Force estimate (Equation 5) \(\hat{\mathbf F} = (C^\top C)^{-1}C^\top\mathbf s\)
Force noise (Equation 6) \(\sigma_s^2(C^\top C)^{-1}\); 120° layout: lateral \(\sqrt{2/3}\,\sigma_s/\beta\), axial \(\sigma_s/(\sqrt3\alpha)\)

Values used in this module

Value Source
Low-voltage area 14.7 vs 20.4 cm²; distinct electrograms 54.4% vs 21.4% Anter et al. (2015)
Interspline spacing 1.5–85 mm; one third of electrodes low voltage Laughner et al. (2016)
Force sensor mean error ≤ 1.0 g over 0–50 g Yokoyama et al. (2008)
Recurrence in 5 of 5 below 10 g; freedom in 8 of 10 above 20 g Reddy et al. (2012)
Heights, spacings, basket radius and layout, sensor noise illustrative
Spacing table, fill distances, force SDs our own code

Interview questions

Try each question before you open its model answer. In a real interview, say which assumptions you are making.

Q1. Why does a small electrode have a higher spreading resistance?

Model answer

Current crowds through a smaller area near the electrode, where most of the resistance lies; for a flush disk it is \(1/(4\sigma a)\), so it scales as one over the radius (Equation 1) (Newman 1966).

Q2. What does spreading resistance cost in recording?

Model answer

Thermal noise grows as the square root of resistance, and a higher source impedance interacts with the amplifier’s input impedance and common-mode rejection (m14).

Q3. A bipole lies along the front’s direction of travel. What does it record, and what if it lies along the front?

Model answer

Along the direction of travel it records full amplitude; along the front itself, ideally nothing, because both electrodes activate together (Equation 2).

Q4. What is the worst-case loss when you take the better of two perpendicular bipoles?

Model answer

\(1 - 1/\sqrt2\), about 29%, when the front travels at 45° to both (Equation 2).

Q5. Why does height above the tissue matter more than electrode size?

Model answer

Height low-pass filters the spatial pattern as \(e^{-kh}\) before any electrode averages it, so it sets the finest detail available (Equation 3); a modeling study found height had the strongest effect (Stinnett-Donnelly et al. 2012).

Q6. Derive the fraction of energy above the spatial Nyquist limit.

Model answer

The power spectrum is \(e^{-2kh}\); integrate from \(\pi/\Delta\) to infinity and divide by the integral from zero: \(e^{-2\pi h/\Delta}\) (Equation 3).

Q7. If the electrodes sit twice as high, what spacing gives the same aliasing?

Model answer

Twice the spacing: \(E\) depends only on \(h/\Delta\) (Equation 3).

Q8. Where are a basket’s electrodes sparsest, and why?

Model answer

At the equator, because splines meet at the poles and the arc between them is \(2\pi R\sin\vartheta/n\) (Equation 4); bunching makes the gaps larger (Laughner et al. 2016).

Q9. How would you compare two basket deployments with the same electrode count?

Model answer

By fill distance over the wall, the fraction of electrodes in contact, and the signal amplitudes; count alone hides gaps (Equation 4).

Q10. Three sensors measure a tip’s deformation. How do you get the force, and what limits its precision?

Model answer

Least squares with a calibration matrix (Equation 5); the covariance \(\sigma_s^2(C^\top C)^{-1}\) shows how noise and layout set precision (Equation 6).

Q11. Why are three sensors 120° apart a good choice?

Model answer

They make \(C^\top C\) diagonal, with equal lateral precision in every direction; among layouts at 0° and ±\(\gamma\), 120° gives the smallest condition number (Equation 6).

Q12. A study reports that higher contact force went with fewer recurrences. What would you want to know?

Model answer

Design (observational or randomized), sample size, how force was averaged, other differences between groups, and the sensor’s accuracy; a 32-patient observational study shows association, not cause (Reddy et al. 2012).

Code lab

The lab computes the design trade-offs of a mapping catheter.

Part What you do Data
A Electrode size and spreading resistance (Equation 1) synthetic
B Bipole direction and orthogonal pairs (Equation 2) synthetic
C Spacing, height and spatial aliasing (Equation 3) synthetic
D Basket coverage (Equation 4) synthetic
E Contact force from three sensors (Equation 5, Equation 6) synthetic

Each part ends with checks: lines that pass with the code as given. If you change something and a check fails, that is information, not an error to hide.

Self-check

Short checks you can do in your head. Open each answer after you have tried it.

S1. A flush disk electrode’s radius halves. Its spreading resistance?

Answer

Doubles (Equation 1).

S2. Disk and hemisphere of the same radius: which has the larger spreading resistance, and by what factor?

Answer

The disk, by \(\pi/2\) (Equation 1).

S3. A bipole at 60° to the direction of travel. Fraction of full amplitude?

Answer

0.5 (Equation 2).

S4. \(h\) = 1 mm and \(\Delta\) = 2 mm. Fraction of energy above Nyquist?

Answer

\(e^{-\pi} \approx\) 4.3% (Equation 3).

S5. 8 splines on a sphere of radius 20 mm. Arc between splines at the equator?

Answer

\(2\pi \times 20/8 \approx\) 15.7 mm (Equation 4).

S6. Sensors 120° apart, noise SD 1, lateral sensitivity 1. SD of the lateral force estimate?

Answer

\(\sqrt{2/3} \approx\) 0.82 (Equation 6).

Literature figures

These figures are worth looking at alongside this module. We link to them instead of copying them: some licenses do not allow reuse here, and others we could not confirm. Each entry says what to look for and where this module explains it.

  • Electrode size and low voltage. Anter and colleagues’ comparison of 3.5-mm and 1-mm electrode maps of the same atria (Anter et al. 2015). This module: spreading resistance and spacing.
  • Resolution against electrode geometry. Stinnett-Donnelly and colleagues’ model of how diameter, length, spacing and height degrade resolution (Stinnett-Donnelly et al. 2012). This module: spacing, height and spatial aliasing.
  • Basket deployment. Laughner and colleagues’ analysis of basket shape, spline spacing and electrode voltage in patients (Laughner et al. 2016). This module: basket geometry.
  • Contact force and lesions. Yokoyama and colleagues’ bench validation and lesion data for an optical force sensor (Yokoyama et al. 2008). This module: contact-force sensing.

Further reading

References

Anter, Elad, Cory M. Tschabrunn, and Mark E. Josephson. 2015. “High-Resolution Mapping of Scar-Related Atrial Arrhythmias Using Smaller Electrodes with Closer Interelectrode Spacing.” Circulation: Arrhythmia and Electrophysiology 8 (3): 537–45. https://doi.org/10.1161/CIRCEP.114.002737.
Deno, D. Curtis, Ram Balachandran, Dennis Morgan, Faisal Ahmad, et al. 2017. “Orientation-Independent Catheter-Based Characterization of Myocardial Activation.” IEEE Transactions on Biomedical Engineering 64 (5): 1067–77. https://doi.org/10.1109/TBME.2016.2589158.
Laughner, Jacob, Shibaji Shome, Nicholas Child, et al. 2016. “Practical Considerations of Mapping Persistent Atrial Fibrillation with Whole-Chamber Basket Catheters.” JACC: Clinical Electrophysiology 2 (1): 55–65. https://doi.org/10.1016/j.jacep.2015.09.017.
Massé, Stéphane, Karl Magtibay, Nicholas Jackson, John Asta, et al. 2016. “Resolving Myocardial Activation with Novel Omnipolar Electrograms.” Circulation: Arrhythmia and Electrophysiology 9 (7): e004107. https://doi.org/10.1161/CIRCEP.116.004107.
Newman, John. 1966. “Resistance for Flow of Current to a Disk.” Journal of the Electrochemical Society 113 (5): 501–2. https://doi.org/10.1149/1.2424003.
Reddy, Vivek Y, Dipen Shah, Josef Kautzner, et al. 2012. “The Relationship Between Contact Force and Clinical Outcome During Radiofrequency Catheter Ablation of Atrial Fibrillation in the TOCCATA Study.” Heart Rhythm 9 (11): 1789–95. https://doi.org/10.1016/j.hrthm.2012.07.016.
Stinnett-Donnelly, Justin M, Nathaniel Thompson, Nicole Habel, et al. 2012. “Effects of Electrode Size and Spacing on the Resolution of Intracardiac Electrograms.” Coronary Artery Disease 23 (2): 126–32. https://doi.org/10.1097/MCA.0b013e3283507a9b.
Yokoyama, Katsuaki, Hiroshi Nakagawa, Dipen C Shah, et al. 2008. “Novel Contact Force Sensor Incorporated in Irrigated Radiofrequency Ablation Catheter Predicts Lesion Size and Incidence of Steam Pop and Thrombus.” Circulation: Arrhythmia and Electrophysiology 1 (5): 354–62. https://doi.org/10.1161/CIRCEP.108.803650.