A low noise current readout architecture for fluorescence detection in living subjects

Abstract

Optical molecular imaging is emerging as a powerful preclinical research tool for investigating and quantifying molecular events in living subjects, with applications including earlier detection of disease, therapeutic monitoring and understanding fundamental biology [1]. For example, imaging the fluorescent molecular probe RGD-Cy5.5, which specifically binds to molecules (α v β 3 integrin receptors) that regulate new blood vessel growth in tumors, can be used to quantify this growth [2]. Capturing the fluorescent signal in living subjects with an implanted biosensor would enable continuous monitoring of tumors in freely moving subjects. Continuous monitoring in the setting of cancer would give valuable information on tumor progression, both in assessing drug efficacy and detecting recurrent tumor growth after treatment. Presently, fluorescence imaging in living subjects is performed with bulky instrumentation that does not permit continuous monitoring of freely moving subjects over long time periods. In order to make a fluorescence-detection system implantable, and portable, a laser excitation source, a photodetector and a readout circuit for measuring and digitizing photocurrents are integrated in a single package, and continuous fluorescence detection is demonstrated in live animals. This paper introduces a readout circuit designed to interface with the fluorescence sensor presented in [3], a monolithically integrated GaAs detector and vertical-cavity surface-emitting laser (VCSEL). The detector signals to be measured are low currents in the range of 5pA to 15nA, with bandwidths up to 100Hz. In order to capture binding dynamics of the fluorescent probe, as well as a wide range of possible tumor sizes and depths, the desired current resolution is 5pA.In applications with such low bandwidths and high sensitivity, a capacitive transimpedance amplifier (CTIA) can be used to provide high SNR through noise averaging [4]. The input current is integrated onto a capacitor C int for a period of time T int . The output of the CTIA is usually sampled and held at the end of T int , before being digitized as a DC signal. The readout noise power introduced at the sampling instant, σ 2 read , limits the maximum achievable SNR of this architecture. If, instead, M samples are taken along the integration ramp and line-fitting is performed, the white noise component of σ 2 read is reduced by M/12, as shown in [5]. Figure 17.5.1 shows our readout system architecture, where ΔΣ modulation is used to acquire and digitize samples taken up the integration ramp. The output of a CTIA is sampled by an incremental ΔΣ modulator, without the need for a sample-and-hold stage at the CTIA output. Before each integration cycle, both the CTIA and the ΔΣ modulator are reset. After reset, the detector current accumulates on C int , giving rise to a voltage ramp at the amplifier's output. This ramp is sampled by the ΔΣ modulator M times during the integration period, such that M=T int f ΣΔ , where f ΣΔ is the sampling frequency of the modulator. The...

Continuous sensing of tumor-targeted molecular probes with a vertical cavity surface emitting laser-based biosensor

“…We are currently designing VCSEL implantable biosensor systems that can be used to interrogate internal tissues and not just externalized tissue. 21 Within our aim to eventually design VCSEL biosensors that operate in freely moving mice, we feel that the fluorescent component of the implanted sensor has now been optimized. To transition to implantable systems, we can now turn our attention to several other design improvements.…”

“…The sensor is potentially applicable to in vitro diagnostics, 20 in vivo exogenously induced fluorescence, 19 and deep tissue fluorescence via implantation. 21 The sensors have a high sensitivity (5 nM in vitro, and autofluorescence-limited to 50 nM in vivo), 19 and because they are produced with standard semiconductor fabrication techniques, the sensors are scalable for both larger pixel formats and lower-cost manufacturing. Thus VCSEL-based biosensors are a good candidate for in vivo continuous molecular sensing.…”

Integrated photonic structures for parallel fluorescence and refractive index biosensing

“…The sensor shows over 3 orders of magnitude of linear operating range, which is an important design parameter in choosing both fluorescence and label-free sensing applications. Shared readout of the detectors for different modality sensors will be aided by signal processing completed in standard CMOS 24 to perform real-time, correlated measurements for repeatable and reliable detection. Noise filtering from such processing will help improve the detection limit and increase the confidence interval for intensity shift measurements.…”

“…With common microfluidic delivery 22,23 and shared electronic read-out, such as the integrated CMOS circuitry we recently demonstrated, 24 one can perform compact, real-time, correlated spectroscopy that ultimately translates to more repeatable and reliable detection.…”

Real-time, continuous, fluorescence sensing in a freely-moving subject with an implanted hybrid VCSEL/CMOS biosensor

“…2) is based on current integration using a capacitive transimpedance amplifier (CTIA), followed by analog-to-digital (A/D) conversion based on sigma-delta (ΣΔ) modulation [15]. The CTIA is used in order to meet the high sensitivity requirement, as it can provide a high signal-to-noise ratio (SNR) through noise averaging.…”