How can engineers improve pulse resolution and cell-type discrimination in modern flow cytometry systems?
Flow cytometry is a technology that analyzes cells from the scatter and fluorescence they produce as they pass through lasers of different wavelengths of light. These signals are detected by optical sensors which produce small electrical pulses of current that are converted to a voltage and digitized. Both the pulse height and pulse area contain information.
Modern flow cytometry systems increasingly require faster event rates, greater than 20,000 events/second, and significantly narrower pulse widths (microsecond-level) to improve cell-size resolution, detect weaker fluorescence signatures, and support multi-parametric analysis with more than 32 channels. These smaller pulse widths place higher demands on the ADC sampling rate, TIA bandwidth, noise performance, and channel density within the acquisition system.
An article titled Flow Cytometer Electronics outlines the challenges for flow cytometry.[1] First, pulses arrive randomly, so to measure the peak height of a 3μs pulse with better than 0.1% accuracy, the pulse envelope needs to be sampled 120 times in those 3μs, implying an ADC sample rate of 40 MSPS. The second challenge is that bright cells may be 10,000x brighter than darker cells, and this calls for a dynamic range of at least four decades of noise-free bits. In the past, cytometers had a myriad of analog signal conditioning electronics, such as integrators and peak detectors, to make up for the fact that ADCs suitable for directly digitizing these pulses were not available at the time.
To summarize, the challenge facing system designers is achieving higher throughput and improved differentiation of cell types. Technically, this requires acquiring narrow detector pulses more accurately as event rates increase in next-generation flow cytometers.
As flow cytometers move toward higher throughput and more fluorescence parameters, the electronics must detect extremely narrow current pulses while maintaining high precision, low 1/f noise, and high linearity over extended integration windows. To meet these application challenges, the component specifications become more demanding:
- ADC sample rates pushed from 15 MSPS → 40 MSPS
- ADC resolution >18 bits, ideally 20 bits, to guarantee sufficient noise-free bits
- Wider bandwidth TIA required to avoid attenuating narrow pulses
- Tighter INL to prevent pulse overlap, or smearing
- Low flicker noise[2] for long-term integration stability
- Very high channel density with low power and low channel-to-channel crosstalk
Analog Devices’ Flow Cytometry Solution
ADI’s flow cytometry solution includes:
- ADA4898-1 TIA front end
- ADA4945-1 fully differential amplifier (FDA)
- AD4080 single-channel SAR ADC (20‑bits, 40 MSPS, ±4 ppm INL)
- LTC6655 low‑noise (3 volt) voltage reference
Figure 1: Fast Precision Flow Cytometry Signal Chain
This solution provides wide TIA bandwidth, stable low‑noise gain and input filtering, and a digitizer capable of accurately sampling narrow microsecond‑level pulses without distortion.
It also delivers immediate performance improvements:
- The flexibility to achieve more samples per pulse (40 MSPS), enabling more accurate cell‑size calculation, or using 8× oversampling to improve system SNR (89.7 dB), supporting detection of weaker fluorescence channels
- Typical ±4 ppm INL (±8 ppm max) improves separation between adjacent pulses
- Lower power and smaller PCB footprint, enabling dense boards with 100+ channels
- Improved integration headroom for future multi-wavelength expansions
The example flow cytometry solution was built in ADI’s Signal Chain Designer Tool. This tool supports drag-and-drop construction of full acquisition signal chain solutions and has analysis capabilities that can be used to assess the relative merits of noise and SNR, power consumption, DC accuracy, frequency response, and solution size. The tool also supports export of the solution to LTspice, where more detailed time and frequency domain simulations can be run. Check out the signal chain solution in Signal Chain Designer: Flow Cytometry signal chain.
Further evaluation of this fast precision optical sensing flow cytometry signal chain is also supported with this LTspice design file.
A pdf version of this Tech Note is also included here.
[1] Snow, C.“. (2004), Flow cytometer electronics.
Cytometry, 57A: 63-69. https://doi.org/10.1002/cyto.a.10120
[2] Castro, Gustavo. (2017). This Noise Will Keep You Up at Night. Analog Dialogue ISSUE #138 https://www.analog.com/en/resources/analog-dialogue/raqs/raq-issue-138.html