Fig. 1. Overall architecture and detailed design of SCAPAS. (a) Layout of the SCAPAS setup. RUA, ring-shaped ultrasonic array; WT, water tank; CN, connector; ZTS, Z-axis translation stage; PL, pump laser; OPO, optical parametric oscillator; LM, laser mirror; BE, beam expander; SR, switchable reflector; BS, beam splitter; PEM, pulse energy meter; RED, rotating engineered diffuser. (b) Laser operation modes for switching between near-infrared and 532 nm outputs. (c) Single-pulse light intensity distribution on the sample surface compared to the distribution after averaging 20 pulses using a randomly rotating diffuser. The speckle effect is visibly suppressed. (d) Single-element configuration and spatial distribution of acoustic field sensitivity (generated by FieldII). (e) Ultrasonic array configuration and the acoustic field sensitivity. (f) Relative positioning of laser illumination, E. coli sample, and array acoustic field during imaging. AF, acoustic field; SP, sample; AB, agarose base. (g) Calibration of lateral resolution.

Fig. 2. Principle of SCAPAS with self-calibration method. (a)–(c) Traditional screening process. (a) Preparation of samples with a single expression. (b) Impact of systematic factors on quantification. RPA, real PA response; SF, systematic factors. (c) Unquantified output using direct readout. (d)–(f) Screening process of SCAPAS. (d) Preparation of E. coli samples co-expressing T-GER and R-GER, with specific requirements for the absorption spectra. (e) Dual-wavelength imaging targeting T-GER and R-GER. (f) Quantified results obtained by eliminating systematic factors using the self-calibration method.

Fig. 3. Image optimization and systematic factor analysis using a numerical calibration sample. (a) Numerical modeling of a single E. coli colony with H and D. (b) Height-encoded map of the calibration sample along the Z axis. (c) Original image reconstructed from 256-channel raw data generated by k-wave simulation. (d) Optimized image. (e) Artifact comparison at the same position before and after optimization, demonstrating its effectiveness in expanding the FOV. (f)–(h) Comparison of colony images, k-space images, and profiles before and after optimization, corresponding to P1, P2, and P3 as labeled in (c) and (d). (i) Settings of R, H, and D values in the calibration sample for the colonies. The radial distribution facilitates the analysis of systematic factors using the control variable method. (j) Variation in PA readouts with D under different R values. (k) Variation in PA readouts with R under different H values. (l) Variation in PA readouts with H under different R values.

Fig. 4. Quantitative simulation results of SCAPAS. (a) Height encoding of numerical samples. Colony morphology and positions are randomly generated within a specified range. (b) Ground truth PA responses of T-GER set across Regions 1 to 4. (c) and (d) Dual-wavelength imaging simulation for T-GER and R-GER, respectively. (e) Direct readout results based on (c), showing a significant difference compared to (b). (f) Quantified results using self-calibration method, in close agreement with the ground truth. (g)–(j) Comparison of PA response obtained using self-calibration and direct readout under different ground truth settings, presented with box plots. In each box, the central mark indicates the median, while the bottom and top edges represent the 25th and 75th percentiles, respectively. (k) Images reconstructed from raw data with different SNRs. (l) Comparison between quantification results of PA response using self-calibration and the ground truth under different SNRs.

Fig. 5. Preliminary experimental results. (a) Absorption spectra of iRFP713, SNIFP, and mScarlet-H. During sample preparation, iRFP713 and SNIFP are used as T-GERs, while mScarlet-H is used as R-GER. (b) Measured signal crosstalk of the three types of FPs, along with the colony without expression (as control), at the selected imaging wavelengths. (c) Bleaching curves of the purified iRFP713 and SNIFP tested in SCAPAS. The imaging duration for the sample is indicated by the blue dashed box. (d) Imaging results of iRFP713 and SNIFP at specific concentrations within the microtubes (averaged over the first 20 frames). (e) Relative PA response ground truth for iRFP713 and SNIFP.

Fig. 6. Quantification results of PA response in E. coli samples using SCAPAS. (a) Dual-wavelength imaging results of E. coli colonies and the workflow for quantifying PA response intensity. The PA responses are normalized to the mean of the corresponding iRFP713 samples for better visualization. (b) PA response distribution maps of the colonies calculated using the direct readout and the self-calibration method, respectively. (c) Box plot presenting the quantitative results using the self-calibration method. In each box, the central mark indicates the median, while the bottom and top edges represent the 25th and 75th percentiles, respectively. The whiskers extend to the furthest data points that are not considered outliers. (d) Box plot of the direct-readout results as a comparison. (e) Intra-group and inter-group T-test matrix based on the quantitative results obtained using the self-calibration method. (f) Intra-group and inter-group T-test matrix based on the readout results.

Fig. 7. Evaluation of the quantification capability of SCAPAS for PA responses. (a) Standard deviation of the quantification results for each sample (iRFP713 samples 1 to 5: 0.212, 0.219, 0.209, 0.342, and 0.289; SNIFP samples 1 to 5: 0.195, 0.214, 0.206, 0.242, and 0.196). The arrows represent the standard deviation of the population comprising all colonies expressing the same T-GER (0.2619 and 0.2185, respectively). (b) Mean of the quantification results for each sample (iRFP713 samples 1 to 5: 4.076, 4.075, 4.083, 4.03, and 4.064; SNIFP samples 1 to 5: 3.273, 3.178, 3.171, 3.264, and 3.216). The arrows represent the mean of the population comprising all colonies expressing the same T-GER (4.065 and 3.228, respectively). (c) Bias matrix showing the deviation of the PA response ratio from the ground truth.

Fig. 8. EIR test in SCAPAS. (a) Schematic diagram of the EIR waveform measurement using edge-emitted signals. (b) Comparison of simulated and measured EIR waveforms. (c) Comparison of simulated and measured EIR spectra.

Fig. 9. Statistical analysis of the height and diameter of actual colonies. (a) Measurement of the average height of actual colonies using the translation focusing method under bright-field microscopy. (b) Colony diameter distribution obtained by fluorescence imaging (using iRFP713 excited at 690 nm as the indicator and observed through a band pass filter with a central wavelength of 710 nm). (c) Comparison of the statistical distributions of the simulated and actual colony diameters.

Fig. 10. Simulation of the influence of deconvolution on quantification results. (a) Reconstructed image with deconvolution. (b) Original reconstructed image. (c) Comparison of quantification results.

Fig. 11. Simulation validation of the influence of colony morphology on quantification results. (a) Numerical sample with dome-shaped colonies. (b) Numerical sample with flat-topped colonies with all other parameters identical to (a). (c) Comparison of quantification results using the self-calibration method for the two groups.

Fig. 12. Light intensity distribution of dual-wavelength illumination. The two distributions are individually normalized, and their difference is then calculated and divided by one of them to obtain the percentage of relative difference.

Fig. 13. Numerical simulation validating the impact of inconsistencies in dual-wavelength illumination on the quantification results. (a) Inconsistent light intensity distribution used in the simulation. Illumination of λT and λR is generated using Gaussian functions with slightly different μ and σ values. (b) Quantification results of PA response intensity under consistent and inconsistent illumination intensity distributions.