Open Access
Add to BibSonomyAbstract
We developed a complementary metal oxide semiconductor (CMOS) integrated device for optogenetic applications. This device can interface via neuronal tissue with three functional modalities: imaging, optical stimulation and electrical recording. The CMOS image sensor was fabricated on 0.35 μm standard CMOS process with built-in control circuits for an on-chip blue light-emitting diode (LED) array. The effective imaging area was 2.0 × 1.8 mm2. The pixel array was composed of 7.5 × 7.5 μm2 3-transistor active pixel sensors (APSs). The LED array had 10 × 8 micro-LEDs measuring 192 × 225 μm2. We integrated the device with a commercial multichannel recording system to make electrical recordings.
©2012 Optical Society of America
1. Introduction
Micro-electrical and mechanical systems (MEMS) and large-scale integration (LSI) technologies have important applications in chemical and biological measurement, such as microflow reactors, chemical synthesis analysis, cell assays, and DNA analysis [1–7]. These devices can efficiently interact with biological microstructures with bumps, cavities, flow, and surface modifications, as well as arrays of sensors and actuators. Complementary metal oxide semiconductor (CMOS) circuits can also be integrated into systems to facilitate amplification and fast read-outs from massively parallel sensors, which directly contact samples, with a high signal-to-noise (S/N) ratio. They also enable the smart activation of microactuators with spatiotemoprally controlled two-dimensional patterns. In the research area of neuroscience and neuroprosthetic devices, examples of such devices include neurochips using ISFETs, on-chip patch clamp recording devices, potentioimaging devices, and silicon probes, which are all established devices in common laboratory use [8–11].
Novel imaging techniques are increasingly and widely used in neuroscience, such as for calcium imaging, voltage sensitive dye (VDS) imaging, intrinsic optical signal (IOS) imaging, and flavoprotein imaging [12–15]. In addition, optical neural stimulation tools, such as channelrhodopsin 2 (ChR2) and halorhodopsin (NpHR) allow the optical stimulation and inhibition of neural cells, respectively, for application in genetically-targeted expression of channel gating proteins. These optogenetic tools are expected to become the fundamental stimulation technique in the field, because they provide several advantages over conventional electrical or chemical stimulation [16, 17]. Photons are the ideal modality for the fine control and monitoring of microstructures, even those within the cells, because their spatial and temporal scale is two to three orders smaller than electrical or chemical phenomena. In principle, it is possible to induce photoisomerizaiton of a retinal molecule using a single photon, which then triggers the gating of a ChR2 membrane protein over a nanosecond timescale. The diffusion or reduction of optical signals is also relatively smaller than other types of signals and they are immune to electromagnetic noise. Another major advantage of optical tools is their compatibility with CMOS technologies. Although array of optical detectors can be directly fabricated in the standard CMOS processes, controlling array of light emitting elements often requires hybrid approach [18].
For these reasons, the development of imaging and optical tools is an urgent challenge for MEMS/CMOS researchers working in the neuroscience. On the imaging side, contact imaging is receiving considerable attention, because it can correct optical signals at the interface of the imaging plane and the biological sample and it does not require the large optics used in microscopy [19, 20]. Micro-LED arrays, silicon probes with polymer light guides, and optorodes have been reportedly used for optical stimulation [21–24]. Despite growing demands, there have been few reports of the development of contact CMOS imagers that are designed for neuroscience applications [25, 26]. To meet the need for compact neuroimaging tools using CMOS technologies, we have developed several types of CMOS image sensors. The first report of these devices was an implantable micro imaging device for the fluorescence detection of enzymatic activities in deep brain structure called hippocampus [27, 28]. The second generation of contact CMOS imaging device was aimed for direct measurement of membrane potentials with VSDs in cultured cortical neurons and in the mouse primary visual cortex [29, 30]. Other type of contact CMOS image sensor utilized near infrared to measure oxygenated state of brain blood to monitor correlated neuronal activity [31]. In this study, we developed a novel CMOS image sensor for optogenetic research. We integrated optical stimulation and electrical recording functions in a CMOS image sensor with an on-chip LED array and a commercial multielectrode array probe.
MEMS and CMOS technologies have potential advantages over current optogenetics experimental setups, which are based on a combination of microscopy, laser scanning and multielectrode arrays, because they can perform massively parallel readouts and activations where the recording/stimulation points number in the thousands or millions. It will also be possible to fully integrate recording and activation functions into an implantable device for use during free-moving in vivo experiments. For chronic implantation experiments, some recent reports of image sensors with energy harvesting function can be also utilized [32].
2. A novel CMOS device for optogenetic applications
2. 1. Device concept
Our novel CMOS device for optogenetic applications incorporates three functions, i.e., static imaging, optical stimulation, and electrical recording. Integrating these three independent functional modalities into one device makes it possible to produce a small multifunctional neural recording and stimulation tool that can be used in simple experimental settings. The imaging function uses a CMOS image sensor for morphological observation and the assignment of stimulation and recording points for any arbitrary biological structures of interest. The CMOS image sensor also has a built-in control circuit, which is integrated with the on-chip LED array. The LED array enables spatiotemporal manipulation of neuronal activities by patterned optical stimulation using a two-dimensional light source array. Electrical recordings are made using a multielectrode array (MEA) probe. The MEA probe is fabricated on a transparent glass substrate, which facilitates transmission of incident/stimulation light to the neural sample. This is achieved by attaching the CMOS device onto the rear of the MEA probe, as shown in Fig. 1 . The integrated microelectrodes allow multichannel recording of the extracellular field potentials produced by neuronal electrical responses.
Fig. 1 Schematic of the multimodal integrated CMOS device used for electrical recording, optical stimulation, and morphological observation of neural tissue.
2. 2. Chip design
The CMOS image sensor was fabricated on 2-poly 4-metal 0.35 μm standard CMOS process (Austria Microsystems). Figure 2 and Table 1 show the chip micrograph and specifications, respectively. The effective recording area of the sensor is 2.0 × 1.8 mm2, with 272 × 240 pixels. The pixel array is composed of 7.5 × 7.5 μm2 3-transistor Active pixel sensors (APSs). In previous work, we developed a CMOS image sensor that bears on-chip microelectrodes for simultaneous contact imaging and electrical recording from a mouse hippocampus slice [33]. The surface of the sensor chip is composed of a passivation layer. However, the passivation layer is removed from the contact microelectrode region, leaving the top Al metal layer exposed. The Al wiring metal layers above the pixel photodiodes are open, facilitating on-chip imaging in the microelectrode region [34]. The 10 × 8 contact electrodes were fabricated using a top metal wiring layer. The electrode is 15 × 15 μm2. The contact electrode and LED drive lines have a two-dimensional lattice structure as shown in [33]. The square aperture in the lattice are the light receptive fields of the photodiodes where the aperture size is 5.6 × 3.1 μm2 (fill factor 31%). This lattice can support 10 × 8 LEDs over the imaging area. The size of the LED is 192 × 225 μm2.
Fig. 2 Chip micrograph of the novel CMOS image sensor and a magnified image of a 2 × 2 LED array and a 3 × 3 pixel array.
Table 1. Specification of CMOS Image Sensor
The CMOS chip circuit diagram and the LED select and drive circuits are shown in Fig. 3a and 3b, respectively. A built-in line scanner and 7-bit decoder are used for addressing the micro-LEDs. One LED is driven by a constant current that is regulated by a transmission gate via on-chip contact electrodes. The scanning mode is used to generate uniform illumination with excitation light during fluorescence imaging, whereas the decoding mode is used for the localized excitation of ChR2 proteins.
2. 3. Device fabrication
On-chip contact electrodes made of Pt thin film were formed on the passivation layer of the CMOS chip as shown in Fig. 4a , using photolithograpy, Pt sputtering, and a lift-off process, as described in previous work [34].
Fig. 4 Micrograph of fabricated on-chip Pt thin film electrodes (200 nm) on the CMOS chip(a), and a magnified image of the anode and cathode LED contact electrodes. Scale bar = 75 μm. (b) Micrograph of the aligned and fixed LED array on the CMOS chip using anisotropic conductive paste (ACP), a 4 × 4 partial region is displayed. Scale bar = 200 μm. (c) Photograph of the completely fabricated device.
The CMOS chip was mounted on the polyimide flexible substrate with epoxy resin and I/O PADs on the chip were connected to Au bonding PADs on flexible printed circuit (FPC) lead wires using Al bonding wires. The bonding wires were mechanically strengthened by reinforcement with ultraviolet curable resin. The electrical resistance between the on-chip Pt thin film electrode and the directly wired I/O PAD was 36.6 (±10.2) Ω. Thus, the 200 nm Pt thin film was continuously deposited onto the passivation layer and the Al layer despite the vertical gap between the two layers (the precise thickness of the passivation layer cannot be disclosed for the reason of confidentiality). A droplet of anisotropic conductive paste (ACP) was casted on the surface of the CMOS chip. An 8 × 10 blue LED array bare chip, measuring 1.8 × 1.9 mm2 was lifted and the electrode plane was placed onto the ACP coated CMOS chip. ACP is a heat-curable resin containing dispersed, 5 μm Ni fillers, which is used in specific three-dimensional integration processes that require the fixation of stacked substrates while retaining electrical conductivity in small regions, such as microelectrodes. The alignment of the LED electrodes to the on-chip contact electrodes was performed manually with microscopic observation. The ACP was thermally cured at 150 °C for 20 s. During the CMOS chip baking, the sapphire substrate was compressed manually to ensure that the Ni fillers were interleaved between the LED electrodes and on-chip electrodes. Figure 4b shows a micrograph of the CMOS chip after thermal fixation of the LED array. The displacement of the two electrodes was several microns at most, while the electrode pitches and shapes were well matched.
In the next step, the multielectrode array probe was attached to the fabricated CMOS chip. The CMOS device was attached to the bottom of the glass substrate using mending tape by aligning the LED/pixel array to the multielectrode array with microscopic observation, after which the peripheral region was fixed using ultraviolet curable resin (Fig. 4c).
3. Functional evaluation
3.1. Morphological observation of a mouse hippocampus slice
We tested a basic use of our imaging device in opto-electrophysiological experiments. It was crucial that the imaging device performed observations of the structural morphology of a biological sample and assigned those structures onto the array of microelectrodes and micro LEDs to yield arbitrary recording and stimulation points. As a proof of concept, we targeted the most commonly used application of planar multielectrode arrays, i.e., the electrophysiological analysis of brain slices from the mouse hippocampus. The mouse hippocampus contains a three-layered structure of pyramidal neurons referred to as CA1, CA3, and dentate gyrus (DG), as shown in Fig. 5a .
Fig. 5 (a) Schematic illustration of the anatomical structure of a mouse hippocampus slice. (b) An image of a hippocampal slice captured using CMOS image sensor. (c) Microscopic observation of a hippocampal slice by optical microscopy. Scale bar = 400 μm.
A mouse hippocampus brain slice was obtained from a male C57BL/6J mouse (aged 6–8 weeks). All procedures were carried out in accordance with the animal care and experimentation guidelines of the Nara Institute of Science and the study was approved by the institutional Animal Care and Use Committee. A prepared brain slice was treated with a fixing solution of 4% paraformaldehyde/phosphate buffered saline (PFA/PBS) overnight. The brain slice was mounted on the multielectrode array chamber, which contained PBS.
Figure 5b shows an image captured by the CMOS image sensor using a completely fabricated version of the device. The effective imaging area was reduced to 50% by the light shielding of the Pt thin film and the LED contact electrodes. It was possible to distinguish hippocampus cell layers in the image captured by the CMOS sensor, as shown in Fig. 5b. For comparison, the micrograph in Fig. 5c was obtained using incident light microscopy in the same experimental conditions. It was possible to assign a particular recording channel that was adjacent to the structural region of interest and to apply multiple photostimulation lights beneath the microelectrode, or at arbitrary positions in the biological microstructures via contact imaging.
The surface of the multielectrode array and the CMOS sensor imaging plane were separated by 1.7 mm in the imaging experiment. The image obtained using this device configuration was different from that acquired by contact imaging. The image of the sample was projected onto the imaging plane by preventing the passage of incident light. By enhancing the linearity of the incident light, it was possible to increase the contrast in the projected image.
3.2. Evaluation of the photostimulation function
According to the LED data sheet, the center of the emission wavelength is 465 nm and the half-bandwidth is 25 nm, which overlaps with the excitation wavelength of ChR2. Figures 6a and 6b show microscopic images of single LED activation at the surface of the sapphire substrate and the glass substrate. The single LED in the array was selected with the built-in decoder circuit in the CMOS image sensor. In Fig. 6b, the illumination area is broader and the contour of LED is blurred because the picture was focused on the microelectrode array. We also conducted a preliminary test of photo-irradiation of a biological sample. As shown in Fig. 6c, photostimulation could be applied to any arbitrary structural region in the hippocampus slice.
Fig. 6 (a) Activated LED in the decoding mode, on the surface of the LED array, and (b) on the MEA probe’s glass substrate. (c) Stimulating light irradiation applied to the mouse hippocampal slice.
The single on-chip LED measured 192 × 225 μm2. A 1.6 mm glass substrate patterned with a Pt black microelectrode array was positioned (MED probe, Alphamed Scientific Inc.) on top of the LED array. We first measured the current luminescence (I–L) characteristics of the on-chip LED and compared it with the transmitted excitation light power on the surface of the glass substrate, when the multielectrode array was attached. The emission power was measured using an optical power meter (TQ8210, Advantest) at 458 nm.
When the 1.6 mm glass substrate was interleaved between the photodetector and the LED array, the emission power was reduced to 54%. The irradiation power measured at the surface of the glass was 104 μW with a drive current of 1.5 mA. It was difficult to determine the power density within the irradiated area simply based on the size of the light spot, because the distance between the LED light source and the photodetector of the optical power meter was 1.6 mm while the size of the detector was 1 cm2. However, we confirmed that the on-chip LEDs were activated with a large drive current of up to 3 mA. The minimum irradiance of light to activate ChR2 is reported between 0.1 and 1 mW/mm2 at 470 nm [20]. It is reasonable to suggest that the on-chip LED array can provide sufficient light power for ChR2 photoactivation from the rear of the glass chamber.
We estimated the emission power distributions, by determining the pixel value profiles of the captured image using fluorescence microscopy (BX 51 and DP 71 CCD, Olympus). Figures 7a and 7b show images captured from an activated LED focused on the surface of the sapphire substrate at different emission power levels, while Figs. 7c and 7d show images captured from the activated LED focused on the glass surface of the MEA. Figures 7e–7h show three-dimensional representations of the pixel values of each image, while Figs. 7i–7l show line profiles obtained in the central section of the captured images.
Fig. 7 (a)~(d) Light emission from single LEDs with different drive currents on the sapphire surface and the glass surface, captured by fluorescence microscopy. (e)~(h) Surface profiles of pixel values, and (i)~(l) line profile of the pixel values in the respective conditions.
It was difficult to accurately monitor the emission power distribution simply based on the captured image when the emission power exceeded 150 μW, because the pixel value was saturated in this high illumination range. Thus, we considered the correlation between the LED emission power and the illumination distribution within the limited power range. In the high power emission range, some of the excitation light was reflected in the sapphire substrate, which broadened the photostimulation spot. The diameter of the photoirradiation spot expanded to 670 μm on the surface of MEA glass substrate when the LED emission power was 43 μW, which presumably triggered action potentials in the ChR2-expressing neurons. The emitted light from micro LEDs also enters to the photodiode array and saturates pixel output. This problem can be evaded by incorporating color filter function to the resin between LED array and CMOS chip, while maintaining anisotropic conductivity.
3.3. Evaluation of the electrical recording function
To validate the other functions of our device, we investigated the electromagnetic coupling of the CMOS image sensor or LED driver circuits with field potential recordings using multielectrode arrays. We used same MEA probe as in the previous section. 50 μm2 Pt black microelectrodes were aligned in an 8 × 8 configuration with a 150 μm pitch. These electrodes were fabricated on a glass substrate patterned using Indium Tin Oxide (ITO) wires with Pt bases. The sapphire substrate of the LED and the CMOS chip that contained the LED array driver and the pixel array readout circuits were placed beneath the recording electrode area, as shown in Fig. 8a .
Fig. 8 (a) Experimental setup of the field potential recoding, (b) image captured using the CMOS image sensor, and field potential recording results in the baseline (c)-(e), artificial signal (f)-(h), and spectral analysis (i)-(k). Scale bar = 400 μm.
The field potential recording experiment was conducted using a multichannel recording system (MED 8 System, Alphamed Scientific Inc.), with a multielectrode array probe, amplifier, low cut and high cut filters, and A/D converter. The gain of the amplifier was 10, while the low cut filter was 0.1 Hz, the high cut frequency was 1 kHz, and the sampling rate was 10 kHz. Figures 8c–8e show time courses of the field potentials in a saline solution recorded from a single microelectrode in different device operation conditions. These operation modes included CMOS image sensor drive mode, LED array drive mode, with halt of circuits in the control experiment. In the LED drive mode, the amplitudes of the base line noise were within 40 μV, which was an acceptable noise level for electrophysiological recording in brain slice experiments [35].
We demonstrated the multimodal functionality of our device by simultaneously operating independent optoelectrical interfaces, as a proof of concept. A 200 Hz/10 mV sinusoid was applied to the saline solution from function generator via W wire electrode as an artificial electrical signal that resembled the extracellular neural responses in terms of frequency element and amplitude. Imaging, photostimulation from the on-chip LED array, and multielectrode recording were performed simultaneously. The captured image and the time course of the recorded field potentials are shown in Figs. 8b and 8f–8h. The tip of the W wire was 100 μm. We observed fine structures on the microelectrode array probe with width measuring tens of microns, such as microelectrodes and ITO wires. The frequency elements of the recorded signal were extracted by spectral analysis in Figs. 8i-8k to investigate the source of EM noise. Most of the artifacts were derived from DC elements that caused base-line fluctuations. In the image sensor drive mode, the power of the high frequency noise was presumably derived from digital circuits, such as the shift register, which accounted for 1.7% of the stimulation signal. These experiments suggest that our device is capable enough to record optically evoked neural activities as extracellular field potentials. To fully validate the functionality of the device from biological perspectives, it will be necessary to conduct electrophysiological experiments using a brain slice of ChR2 expressing transgenic mice.
These results support the feasibility of our device for morphological observation, spatiotemporal control of photo-sensitized neural tissue, and extracellular independent recordings using the CMOS image sensor, on-chip LED array, and conventional multielectrode array. Table 2 summarizes the figure of merits of the device for each functional validation.
Table 2. Summary of Device Performance
4. Conclusions
We designed and fabricated a novel CMOS integrated device for opotogenetic applications. We demonstrated three functional modalities of our device, as follows: morphological observations of a mouse hippocampus brain slice using the CMOS image sensor; applying a blue stimulation light using an LED array controlled by built-in CMOS circuits; and field potential recordings of an artificial signal using a multielectrode array probe, as a feasibility study. These results pave the way for building CMOS-based optoelectronic neural interfaces that can fully utilize the benefits of LSI and MEMS technologies in terms of massively parallel recording and stimulation.
Acknowledgments
This work was supported by the Core Research for Evolutional Science and Technology (CREST) project of the Japan Science and Technology Agency (JST) and by the Japan Society for the Promotion of Science (JSPS).
References and links
1. M. W. Pruessner, T. H. Stievater, M. S. Ferraro, W. S. Rabinovich, J. L. Stepnowski, and R. A. McGill, “Waveguide micro-opto-electro-mechanical resonant chemical sensors,” Lab Chip 10(6), 762–768 (2010). [CrossRef] [PubMed]
2. A. Hierlemann and H. Baltes, “CMOS-based chemical microsensors,” Analyst (Lond.) 128(1), 15–28 (2003). [CrossRef] [PubMed]
3. C. J. Lu, W. H. Steinecker, W. C. Tian, M. C. Oborny, J. M. Nichols, M. Agah, J. A. Potkay, H. K. Chan, J. Driscoll, R. D. Sacks, K. D. Wise, S. W. Pang, and E. T. Zellers, “First-generation hybrid MEMS gas chromatograph,” Lab Chip 5(10), 1123–1131 (2005). [CrossRef] [PubMed]
4. T. Tokuda, H. Yamada, K. Sasagawa, and J. Ohta, “Polarization-analyzing CMOS image sensor with monolithically embedded polarizer for microchemistry systems,” IEEE Trans. Biomed. Circuits Syst. 3(5), 259–266 (2009). [CrossRef]
5. P. C. Chen, Y. Y. Huang, and J. L. Juang, “MEMS microwell and microcolumn arrays: novel methods for high-throughput cell-based assays,” Lab Chip 11(21), 3619–3625 (2011). [CrossRef] [PubMed]
6. G. Baaken, M. Sondermann, C. Schlemmer, J. Rühe, and J. C. Behrends, “Planar microelectrode-cavity array for high-resolution and parallel electrical recording of membrane ionic currents,” Lab Chip 8(6), 938–944 (2008). [CrossRef] [PubMed]
7. T. Tokuda, I. Kadowaki, K. Kagawa, M. Nunoshita, and J. Ohta, “A new scheme for imaging on-chip dry DNA spots using optical/potential dual-image complementary metal oxide semiconductor sensor,” Jpn. J. Appl. Phys. 46(4B), 2806–2810 (2007). [CrossRef]
8. B. Eversmann, M. Jenkner, F. Hofmann, C. Paulus, R. Brederlow, B. Holzapfl, P. Fromherz, M. Merz, M. Brenner, M. Schreiter, R. Gabl, K. Plehnert, M. Steinhauser, G. Eckstein, D. Schmitt-Landsiedel, and R. Thewes, “A 128 x 128 CMOS biosensor array for extracellular recording of neural activity,” IEEE J. Solid-state Circuits 38(12), 2306–2317 (2003). [CrossRef]
9. P. Weerakoon, E. Culurciello, Y. Yang, J. Santos-Sacchi, P. J. Kindlmann, and F. J. Sigworth, “Patch-clamp amplifiers on a chip,” J. Neurosci. Methods 192(2), 187–192 (2010). [CrossRef] [PubMed]
10. K. Imfeld, S. Neukom, A. Maccione, Y. Bornat, S. Martinoia, P. A. Farine, M. Koudelka-Hep, and L. Berdondini, “Large-scale, high-resolution data acquisition system for extracellular recording of electrophysiological activity,” IEEE Trans. Biomed. Eng. 55(8), 2064–2073 (2008). [CrossRef] [PubMed]
11. R. J. Vetter, J. C. Williams, J. F. Hetke, E. A. Nunamaker, and D. R. Kipke, “Chronic neural recording using silicon-substrate microelectrode arrays implanted in cerebral cortex,” IEEE Trans. Biomed. Eng. 51(6), 896–904 (2004). [CrossRef] [PubMed]
12. C. Stosiek, O. Garaschuk, K. Holthoff, and A. Konnerth, “In vivo two-photon calcium imaging of neuronal networks,” Proc. Natl. Acad. Sci. U.S.A. 100(12), 7319–7324 (2003). [CrossRef] [PubMed]
13. I. Ferezou, S. Bolea, and C. C. H. Petersen, “Visualizing the cortical representation of whisker touch: voltage-sensitive dye imaging in freely moving mice,” Neuron 50(4), 617–629 (2006). [CrossRef] [PubMed]
14. R. D. Frostig, E. E. Lieke, D. Y. Ts’o, and A. Grinvald, “Cortical functional architecture and local coupling between neuronal activity and the microcirculation revealed by in vivo high-resolution optical imaging of intrinsic signals,” Proc. Natl. Acad. Sci. U.S.A. 87(16), 6082–6086 (1990). [CrossRef] [PubMed]
15. K. C. Reinert, R. L. Dunbar, W. Gao, G. Chen, and T. J. Ebner, “Flavoprotein autofluorescence imaging of neuronal activation in the cerebellar cortex in vivo,” J. Neurophysiol. 92(1), 199–211 (2004). [CrossRef] [PubMed]
16. B. R. Arenkiel, J. Peca, I. G. Davison, C. Feliciano, K. Deisseroth, G. J. Augustine, M. D. Ehlers, and G. Feng, “In vivo light-induced activation of neural circuitry in transgenic mice expressing channelrhodopsin-2,” Neuron 54(2), 205–218 (2007). [CrossRef] [PubMed]
17. V. Gradinaru, M. Mogri, K. R. Thompson, J. M. Henderson, and K. Deisseroth, “Optical deconstruction of parkinsonian neural circuitry,” Science 324(5925), 354–359 (2009). [CrossRef] [PubMed]
18. A. L. Lentine, K. W. Goossen, J. A. Walker, L. M. F. Chirovsky, L. A. D’Asaro, S. P. Hui, B. J. Tseng, R. E. Leibenguth, J. E. Cunningham, W. Y. Jan, J. Kuo, D. W. Dahringer, D. P. Kossives, D. D. Bacon, G. Livescu, R. L. Morrison, R. A. Novotny, and D. B. Buchholz, “High-speed optoelectronic VLSI switching chip with >4000 optical I/O based on flip-chip bonding of MQW modulators and detectors to silicon CMOS,” IEEE J. Sel. Top. Quantum Electron. 2(1), 77–84 (1996). [CrossRef]
19. X. Cui, L. M. Lee, X. Heng, W. Zhong, P. W. Sternberg, D. Psaltis, and C. Yang, “Lensless high-resolution on-chip optofluidic microscopes for Caenorhabditis elegans and cell imaging,” Proc. Natl. Acad. Sci. U.S.A. 105(31), 10670–10675 (2008). [CrossRef] [PubMed]
20. J. Honghao, P. A. Abshire, M. Urdaneta, and E. Smela, “CMOS contact imager for monitoring cultured cells,” in Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS)4, 3491–3494 (2005).
21. N. Grossman, V. Poher, M. S. Grubb, G. T. Kennedy, K. Nikolic, B. McGovern, R. B. Palmini, Z. Gong, E. M. Drakakis, M. A. A. Neil, M. D. Dawson, J. Burrone, and P. Degenaar, “Multi-site optical excitation using ChR2 and micro-LED array,” J. Neural Eng. 7(1), 016004 (2010). [CrossRef] [PubMed]
22. M. Im, I. Cho, F. Wu, K. D. Wise, and E. Yoon, “Neural probes integrated with optical mixer/splitter waveguides and multiple stimulation sites,” in Proceedings of IEEE International Conference on Micro Electro Mechanical Systems (MEMS) (Cancun, Mexico, 2011), pp. 1051–1054.
23. R. Kobayashi, S. Kanno, S. Sasaki, S. Lee, M. Koyanagi, H. Yao, and T. Tanaka, “Development of Si neural probe with optical waveguide for highly accurate optical stimulation of neuron,” in Proceedings of IEEE International EMBS Conference on Neural Engineering (Cancun, Mexico, 2011), pp. 294–297.
24. J. Zhang, F. Laiwalla, J. A. Kim, H. Urabe, R. Van Wagenen, Y. K. Song, B. W. Connors, F. Zhang, K. Deisseroth, and A. V. Nurmikko, “Integrated device for optical stimulation and spatiotemporal electrical recording of neural activity in light-sensitized brain tissue,” J. Neural Eng. 6(5), 055007 (2009). [CrossRef] [PubMed]
25. F. Normandin, M. Sawan, and J. Faubert, “A new integrated front-end for a noninvasive brain imaging system based on near-infrared spectroreflectometry,” IEEE Trans. Circuits. Syst., l. Regul. Pap. 52(12), 2663–2671 (2005). [CrossRef]
26. J. H. Park, V. Pieribone, K. Dongsoo, J. V. Verhagen, C. von Hehn, and E. Culurciello, “High-speed fluorescence imaging system for freely moving animals,” in Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS) (Taipei, Taiwan, 2009), pp. 2429–2432.
27. D. C. Ng, H. Tamura, T. Mizuno, T. Tokuda, M. Nunoshita, Y. Ishikawa, S. Shiosaka, and J. Ohta, “An implantable and fully integrated complementary metal-oxide semiconductor device for in vivo neural imaging and electrical interfacing with the mouse hippocampus,” Sens. Actuators A Phys. 145–146, 176–186 (2008). [CrossRef]
28. H. Tamura, D. C. Ng, T. Tokuda, H. Naoki, T. Nakagawa, T. Mizuno, Y. Hatanaka, Y. Ishikawa, J. Ohta, and S. Shiosaka, “One-chip sensing device (biomedical photonic LSI) enabled to assess hippocampal steep and gradual up-regulated proteolytic activities,” J. Neurosci. Methods 173(1), 114–120 (2008). [CrossRef] [PubMed]
29. T. Kobayashi, A. Tagawa, T. Noda, K. Sasagawa, T. Tokuda, Y. Hatanaka, H. Tamura, Y. Ishikawa, S. Shiosaka, and J. Ohta, “Potentiometric dye imaging for pheochromocytoma and cortical neurons with a novel measurement system using and integrated complementary metal-oxide-semiconductor imaging device,” Jpn. J. Appl. Phys. 49(11), 117001 (2010). [CrossRef]
30. T. Kobayashi, H. Tamura, Y. Hatanaka, M. Motoyama, T. Noda, K. Sasagawa, T. Tokuda, Y. Ishikawa, S. Shiosaka, and J. Ohta, “Functional neuroimaging by using an implantable CMOS multimodal device in a freely-moving mouse” in Proceedings of IEEE Biomedical Circuits and Systems Conference (BioCAS) (San Diego, USA, 2011), pp. 110–113.
31. S. Shishido, Y. Oguro, T. Noda, K. Sasagawa, T. Tokuda, and J. Ohta, “CMOS image sensor for recording of intrinsic-optical-signal of the brain,” in Proceedings of IEEE International SoC Design Conference (ISOCC) (Busan, Korea, 2009), pp. 190–193.
32. C. Shi, M. K. Law, and A. Bermak, “A novel asynchronous pixel for an energy harvesting CMOS image sensor,” IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 19(1), 118–129 (2011). [CrossRef]
33. A. Tagawa, H. Minami, M. Mitani, T. Noda, K. Sasagawa, T. Tokuda, H. Tamura, Y. Hatanaka, Y. Ishikawa, S. Shiosaka, and J. Ohta, “Multimodal complementary metal-oxide-semiconductor sensor device for imaging of fluorescence and electrical potential in deep brain of mouse,” Jpn. J. Appl. Phys. 49(1), 01AG02 (2010). [CrossRef]
34. A. Nakajima, T. Noda, K. Sasagawa, T. Tokuda, Y. Ishikawa, S. Shiosaka, and J. Ohta, “Planar multielectrode array coupled complementary metal oxide semiconductor image sensor for in vitro electophysiology,” Jpn. J. Appl. Phys. 50(4), 04DL04 (2011). [CrossRef]
35. M. O. Heuschkel, M. Fejtl, M. Raggenbass, D. Bertrand, and P. Renaud, “A three-dimensional multi-electrode array for multi-site stimulation and recording in acute brain slices,” J. Neurosci. Methods 114(2), 135–148 (2002). [CrossRef] [PubMed]
