Standard ADC components seldom provide enough flexibility to meet the conflicting needs of today’s medical designs
By Ian Lankshear,
Electrocardiography (ECG) plays a central role in both diagnostic and ongoing health monitoring. This is especially true for wearable and implantable medical devices as they become more mainstream.
As device requirements for power consumption, size, and signal quality grow stricter, many engineers are turning to application-specific integrated circuits (ASICs) to meet these demands. Within these ASICs, the analog-to-digital converter (ADC) must be carefully optimized for power efficiency, resolution, and noise control.
In the first article of this two-part series, we’ll offer an overview of ECG systems, lead generation and how channel count affects ADC selection. In the second article, we’ll look at the principal trade-offs engineers face when choosing and refining ADCs for ECG ASIC design, providing context for tailoring these parameters to each use case.
ECG waveform characteristics
An ECG records the heart’s electrical activity, mapping out depolarization and repolarization events in the atria and ventricles. These signals tend to be quite subtle, as most frequencies are under 150 Hz.
A standard ECG waveform has several distinct segments, each offering its own clinical insights. The P wave shows atrial depolarization. It is small (<0.25 mV) and low frequency. The QRS complex indicates ventricular depolarization and includes Q wave (initial negative deflection), R wave (the tallest, positive spike), and the S wave (negative deflection following the R wave — this segment carries most of the waveform’s energy and the highest frequency content, up to 150 Hz). Finally, the T wave (representing ventricular repolarization) is broader and smoother than the QRS complex.
This illustration of a standard ECG waveform shows labelled intervals that serve as reference points in clinical and veterinary practice.
Intervals and durations such as the PR interval, QRS duration, and QT interval are key diagnostic markers.
The specific shape of an ECG means that much of the clinically valuable information is found in short, well-defined periods. This makes adaptive sampling a practical strategy for conserving power, as most of the signal’s energy is concentrated around the QRS complex. The remainder of the ECG trace tends to be flatter, with low-frequency sections.
Leads, channels and electrode placement
To make sense of an ECG system, it helps to understand the difference between electrodes, leads, and channels.
Electrodes are the physical contact points placed on the skin. These stickers, clips or patches pick up the tiny voltage changes caused by the heart’s electrical activity.
Leads are representations of the heart’s activity, formed by comparing electrical signals from two or more electrodes. A lead is not an electrode itself, but a derived voltage difference. In humans, the classic Einthoven’s triangle is formed by Lead I (left arm minus right arm), Lead II (left leg minus right arm), and Lead III (left leg minus left arm). In veterinary medicine, the same approach applies, though electrode placement is adapted to suit different animal anatomies, such as dogs recorded while lying on their side or horses using a base-apex configuration.
Channels are the output traces generated by the ECG system. Each channel corresponds to one lead shown on the screen or paper. For example, a three-channel Holter monitor might display three leads at once (e.g., II, V1, V5). A 12-lead ECG uses 10 electrodes but computes 12 distinct leads mathematically and displays them across 12 channels.
These distinctions help clarify how raw electrode signals become clinically meaningful ECG traces and highlight the importance of clear terminology and system design.
Lead generation methods
Different approaches can be used to generate ECG leads, each providing a slightly different view of cardiac activity:
- Bipolar leads (Einthoven’s) measure potential differences between limb electrodes.
- Augmented leads (Goldberger) use a single limb electrode referenced against the average of the other two.
- Precordial leads (Wilson) use chest electrodes referenced to a limb average and are adapted for different species in veterinary applications.
- Special leads include vectorcardiographic or base–apex configurations, particularly for large animals.
For ASIC designers, the number of ADC channels determines hardware complexity, while the number of leads is a matter of signal processing and mathematical recombination. Efficient multiplexing and fast ADC settling are essential for reconstructing multiple simultaneous leads when channel availability is limited.
Impact of channel count on ADC selection
The number of channels in an ECG system directly influences the ADC requirements.
In single- or dual-lead systems such as wearables and fitness monitors, one or two ADC channels are typically sufficient, requiring moderate resolution (10–12 bits) and ultra-low power operation. As channel count increases, for example, in five-lead telemetry or full 12-lead diagnostic ECG, the system complexity grows considerably.
More channels require either several ADCs running in parallel or a single ADC capable of high-speed multiplexing with minimal settling delay. High-resolution incremental sigma-delta (ΣΔ) ADCs are especially well suited to this task, offering flexible trade-offs between noise performance and conversion time. This makes it possible to digitize each channel under optimized conditions using just one ADC core.
For advanced applications such as diagnostic ECG or fetal ECG, where signal amplitudes may differ by more than an order of magnitude, the ADC needs to provide both a high effective number of bits (ENOB) and a broad dynamic range to capture small signal details accurately and without distortion.
Since both power consumption and data bandwidth increase with the number of channels, modern ECG ASICs tend to integrate front-end filtering and on-chip digital signal processing (DSP) that compress or preprocess signals before transmission.
Choosing a suitable ADC architecture — one that aligns with the electrode configuration, resolution, and sampling rate — remains fundamental for achieving the best mix of performance, energy efficiency and silicon area in multi-channel ECG designs.
Read more about that in our second article.
This article has first been published at www.medicaldesignandoutsourcing.com on December 2, 2025.
