Search

SIGNIFICANCE: We present a motion-resistant three-wavelength spatial frequency domain imaging (SFDI) system with ambient light suppression using an 8-tap complementary metal-oxide semiconductor (CMOS) image sensor (CIS) developed at Shizuoka University. The system addresses limitations in conventional SFDI systems, enabling reliable measurements in challenging imaging scenarios that are closer to real-world conditions. AIM: Our study demonstrates a three-wavelength SFDI system based on an 8-tap CIS. We demonstrate and evaluate the systems capability of mitigating motion artifacts and ambient light bias through tissue phantom reflectance experiments and in vivo volar forearm experiments. APPROACH: We incorporated the Hilbert transform to reduce the required number of projected patterns per wavelength from three to two per spatial frequency. The 8-tap image sensor has eight charge storage diodes per pixel; therefore, simultaneous image acquisition of eight images based on multi-exposure is possible. Taking advantage of this feature, the sensor simultaneously acquires images for planar illumination, sinusoidal pattern projection at three wavelengths, and ambient light. The ambient light bias is eliminated by subtracting the ambient light image from the others. Motion artifacts are suppressed by reducing the exposure and projection time for each pattern while maintaining sufficient signal levels by repeating the exposure. The system is compared to a conventional SFDI system in tissue phantom experiments and then in vivo measurements of human volar forearms. RESULTS: The 8-tap image sensor-based SFDI system achieved an acquisition rate of 9.4 frame sets per second, with three repeated exposures during each accumulation period. The diffuse reflectance maps of three different tissue phantoms using the conventional SFDI system and the 8-tap image sensor-based SFDI system showed good agreement except for high scattering phantoms. For the in vivo volar forearm measurements, our system successfully measured total hemoglobin concentration, tissue oxygen saturation, and reduced scattering coefficient maps of the subject during motion (16.5 cm/s) and under ambient light (28.9 lx), exhibiting fewer motion artifacts compared with the conventional SFDI. CONCLUSIONS: We demonstrated the potential for motion-resistant three-wavelength SFDI system with ambient light suppression using an 8-tap CIS.

We demonstrate a motion-resistant, three-wavelength, spatial frequency domain imaging (SFDI) system with ambient light suppression using a new 8-tap CMOS image sensor developed in our laboratory. Compared to the previous sensor (134×150), the new sensor’s readout maximum frame rate has improved to 33fps from 6.28fps, and the new 700×540-pixel sensor allows imaging at a higher spatial resolution over a larger field of view. Furthermore, the number of projected images needed per wavelength is reduced from three to two after applying the Hilbert transform. One image of planar illumination and one image for sinusoidal pattern projection at three wavelengths as well as one image of ambient light are captured by the 8-tap image sensor concurrently. The bias caused by ambient light is removed by subtracting the ambient light image from other images. Suppression of motion artifacts is achieved by reducing the exposure and projection time of each pattern. Sufficient signal level is maintained by repeating the exposure multiple times. In this study, LEDs with wavelengths of 554nm, 660nm, and 730nm were used to estimate oxy-/deoxyhemoglobin and melanin concentrations from in-vivo volar forearm skin.

We have developed a motion-resistant three-wavelength spatial frequency domain imaging (SFDI) system using an 8-tap CMOS image sensor. We have successfully measured two wavelengths using the same sensor. However, to remove the effect of melanin, measuring at three wavelengths is needed. In this study, the Hilbert transform is introduced to reduce the number of captured images for each wavelength from three to two. In the experiments, a DMD was illuminated with LEDs with wavelengths of 660nm, 780nm, and 850nm. One plane and one sinusoidal pattern were projected on the specimen for each wavelength. In addition, one ambient light image was captured to extract only the reflection components for the projected patterns. Pattern projection and exposure were repeated multiple times at a faster frame rate than the video rates to suppress the motion artifact while keeping appropriate signal levels. Totally 7 images were captured at once. Using this system, we demonstrated suppression of motion artifacts and ambient light in capturing a moving wrist.

We are developing a method for simultaneous multi-band spatial frequency domain imaging (SFDI) and blood flow mapping by multi-exposure laser speckle contrast imaging (MELSCI) with a laboratory-designed 2x2-aperture 4-tap CMOS image sensor. Four-tap CMOS image sensor can capture four images synchronized with controlled illuminations and can implement flexible exposure. Although we have demonstrated SFDI with suppressing motion artifact and ambient light bias and efficient video-rate MELSCI based on a four-tap CMOS image sensor separately, simultaneous multi-modal imaging is effective for improving the accuracy of medical diagnosis. The image sensing area is divided into 2×2 regions, and each region is equipped with optical filters and an imaging lens. By using optical bandpass filters, multi-wavelength imaging and wavelength-division multiplexed imaging are realized. Using the fabricated image sensor, we observed a human wrist. 450, 550, 660nm LEDs were used for 3-band SFDI, and a 785nm LD was used for MELSCI. In order to measure a change in the blood flow speed, an examinee was measured before and after an exercise (squatting 30 times). The unit exposure duration for SFDI and MELSCI was 10ms. The pattern was generated by a DMD. The spatial frequency of the projected sinusoidal patterns was 0.1mm-1. MDC and MAC maps for the three wavelengths by SFDI, and K2 value maps by MELSCI were successfully obtained.

Insufficient signal-to-noise ratio (SNR) limits the quality of cardiac pulses measured by multitap CMOS image sensors (CISs). We present an in-pixel temporal redundant samplings (TRS) method for enhancing the SNR through sampling the charges originated by one individual optical scene into the multiple storage taps in turn with the identical time window, enabling magnification of the image signal in shot noise domain. Modeling indicates that the linear SNR increases with a factor square root of TRS sampling number, which is also verified by characterization results utilizing a four-tap CIS. Application results show that the high-frequency (HF) noise components observed in the measured cardiac pulses can be attenuated by 67.9%, 48%, and 53.6% with heart rate detection accuracies up to 99.7%, 99.3%, and 98.7% by applying TRS modes under given conditions of stable ambient, background light (BGL) variation, and motion disturbance, respectively. Moreover, the proposed method is as well foreseeable to be extended to other multitap-involved imaging applications. [ABSTRACT FROM AUTHOR]

In typical stimulated Raman scattering (SRS) signal extraction, the photodetector and lock-in amplifier are often based on separate platforms, rendering the system cumbersome and non-scalable. This paper proposes an SRS double-demodulation lock-in amplifier implemented with a complementary metal-oxide semiconductor (CMOS) image sensor technology that integrates two-stage 1/f noise and offset reduction circuits with a high-speed lateral electric field modulation (LEFM) photo-demodulator. A weak SRS signal is buried in a large offset with a ratio of 10−4 to 10−6; boosting such signals in a CMOS device requires an extremely high offset and noise reduction capability. The double-modulation two-stage lock-in amplifier demodulates at 40 MHz with a sampling frequency of 20 MHz, can suppress the laser and circuit's 1/f noise to achieve higher detection sensitivity. A prototype chip fabricated using 0.11 μm CMOS image sensor technology is evaluated. Both simulation and measurement results are presented to verify the functionality and show that the differential readout structure can successfully reject laser common mode components while emphasizing its differences. The measurement results show that the double-modulation lock-in amplifier effectively suppresses the circuit's 1/f noise by a factor of nearly two decades. [ABSTRACT FROM AUTHOR]

In robot vision, it is often desired to measure an area larger than the field of view of the camera, so the camera tends to be mounted on a mechanical pan/tilt platform. However, such mechanisms have a non-negligible response time compared to the frame rate of the camera. In this paper, we describe what we believe to be a new method that allows arbitrary and multiple gaze directions to be observed in a frame-by-frame manner based on a resonant mirror and a lock-in pixel image sensor. In the proposed method, while the gaze direction oscillates due to the resonance mirror, the image sensor makes an exposure of several hundreds of nanoseconds every time the gaze passes through the direction to be captured, and accumulates the captured signals. A prototype system was developed using a lock-in pixel image sensor with four image storage units called taps and a resonant mirror with a resonant frequency of 12 kHz. The system achieved both arbitrary control of the gaze direction in a frame-by-frame manner, and simultaneous capturing of four images with arbitrary gaze directions also in a frame-by-frame manner.