A denoising model based on multi-agent reinforcement learning with data transformation for digital tomosynthesis.

Objective. Denoising models based on the supervised learning have been proposed for medical imaging. However, its clinical availability in digital tomosynthesis (DT) imaging is limited due to the necessity of a large amount of training data for providing acceptable image quality and the difficulty in minimizing a loss. Reinforcement learning (RL) can provide the optimal pollicy, which maximizes a reward, with a small amount of training data for implementing a task. In this study, we presented a denoising model based on the multi-agent RL for DT imaging in order to improve the performance of the machine learning-based denoising model.Approach. The proposed multi-agent RL network consisted of shared sub-network, value sub-network with a reward map convolution (RMC) technique and policy sub-network with a convolutional gated recurrent unit (convGRU). Each sub-network was designed for implementing feature extraction, reward calculation and action execution, respectively. The agents of the proposed network were assigned to each image pixel. The wavelet and Anscombe transformations were applied to DT images for delivering precise noise features during network training. The network training was implemented with the DT images obtained from the three-dimensional digital chest phantoms, which were constructed by using clinical CT images. The performance of the proposed denoising model was evaluated in terms of signal-to-noise ratio (SNR), structural similarity (SSIM) and peak signal-to-noise ratio (PSNR).Main results. Comparing the supervised learning, the proposed denoising model improved the SNRs of the output DT images by 20.64% while maintaining the similar SSIMs and PSNRs. In addition, the SNRs of the output DT images with the wavelet and Anscombe transformations were 25.88 and 42.95% higher than that for the supervised learning, respectively.Significance. The denoising model based on the multi-agent RL can provide high-quality DT images, and the proposed method enables the performance improvement of machine learning-based denoising models.

Comparison of Macrophage Immune Responses and Metabolic Reprogramming in Smooth and Rough Variant Infections of Mycobacterium mucogenicum.

Mycobacterium mucogenicum (Mmuc), a rapidly growing nontuberculous mycobacterium (NTM), can infect humans (posttraumatic wound infections and catheter-related sepsis). Similar to other NTM species, Mmuc exhibits colony morphologies of rough (Mmuc-R) and smooth (Mmuc-S) types. Although there are several case reports on Mmuc infection, no experimental evidence supports that the R-type is more virulent. In addition, the immune response and metabolic reprogramming of Mmuc have not been studied on the basis of morphological characteristics. Thus, a standard ATCC Mmuc strain and two clinical strains were analyzed, and macrophages were generated from mouse bone marrow. Cytokines and cell death were measured by ELISA and FACS, respectively. Mitochondrial respiration and glycolytic changes were measured by XF seahorse. Higher numbers of intracellular bacteria were found in Mmuc-R-infected macrophages than in Mmuc-S-infected macrophages. Additionally, Mmuc-R induced higher levels of the cytokines TNF-α, IL-6, IL-12p40, and IL-10 and induced more BMDM necrotic death. Furthermore, our metabolic data showed marked glycolytic and respiratory differences between the control and each type of Mmuc infection, and changes in these parameters significantly promoted glucose metabolism, extracellular acidification, and oxygen consumption in BMDMs. In conclusion, at least in the strains we tested, Mmuc-R is more virulent, induces a stronger immune response, and shifts bioenergetic metabolism more extensively than the S-type. This study is the first to report differential immune responses and metabolic reprogramming after Mmuc infection and might provide a fundamental basis for additional studies on Mmuc pathogenesis.

Effect of smoking on clinical outcomes in patients receiving rotational atherectomy in calcified coronary lesions: from the ROCK Registry, South Korea.

BACKGROUND: Tobacco smoking and its harmful health effects also increase economic burdens globally. Surprisingly, despite the detrimental health consequences of smoking, some studies have shown better survival among smokers compared with non-smokers, a phenomenon called "smoker's paradox". However, the impact of smoking status on clinical outcomes in severe calcified coronary artery disease (CAD) patients has yet to be reported. OBJECTIVES: Investigate the impact of smoking on clinical outcomes in calcified CAD receiving rotational atherectomy (RA). DESIGN: Retrospective review of medical records. SETTING: Multicenter registry in South Korea. PATIENTS AND METHODS: This multicenter registry included consecutive patients with calcified CAD who underwent RA at nine tertiary centers in Korea between January 2010 and October 2019. MAIN OUTCOME MEASURES: Target-vessel failure (TVF) which included the composite of cardiac death, target-vessel myocardial infarction (TVMI), and target-vessel revascularization (TVR). SAMPLE SIZE: 583 lesions in 540 patients followed for a median of 16.1 months. RESULTS: Lesions were divided into two groups: non-smokers (n=472, 81.0%) and smokers (n=111, 19.0%). TVF in the smoker group was significantly more frequent than in non-smoker group (log rank P=.016). The inverse probability of treatment weighting analysis also showed that smoking was significantly associated with a higher incidence of the primary outcome (HR: 1.617; 95% CI: 1.127-2.320; P=.009), cardiac death (HR 1.912; 95% CI: 1.105-3.311; P=.021), myocardial infarction (HR: 3.914; 95% CI: 1.884-8.132; P<.001), TVMI (HR: 3.234; 95% CI: 1.130-9.258; P=.029), and TVR (HR: 1.661; 95% CI: 1.043-2.643; P=.032). However, any bleeding was significantly observed less in the smokers. CONCLUSION: Smoking is significantly associated with adverse clinical outcomes in CAD patients requiring RA. LIMITATIONS: Retrospective design. CONFLICTS OF INTEREST: None.

Evaluation of a new foetal shielding device for pregnant brain tumour patients.

BACKGROUND: The present study aimed to propose a new foetal shielding device for pregnant cancer patients to reduce the foetal dose associated with treatment techniques using multiple gantry angles, such as intensity-modulated radiation therapy (IMRT) or volumetric modulated arc therapy (VMAT). METHODS: Three shielding structures were designed to minimise the scattered and leaked radiation from various gantry angles and radiation scattering within the patient. The base-plate part that can be placed on the treatment couch was designed to reduce the scattered and leaked radiation generated at gantry angles located near 180°. A body shielding part that can cover the lower chest and abdomen was designed, and a neck-shielding structure was added to reduce the internal and external radiation scattering from the treatment area. Evaluation plans were generated to assess the foetal dose reduction by the foetal shielding device in terms of the shielding material thickness, distance from the field edge, and shielding component using the flattened 6 MV photon beam (6MV) and flattening filter-free 6 MV photon beam (6MV-FFF). In addition, the effectiveness of the foetal shielding device was evaluated in a pregnant brain tumour patient. RESULTS: The shielding material consisting of three parts was placed on frames composed of four arch shapes with a vertical curved structure, connection bar at the top position, and base plate. Each shielding part resulted in reductions in the radiation dose according to the treatment technique, as the thickness of the shielding material increased and the foetal dose decreased. In addition, a foetal dose reduction of approximately 50% was confirmed at 50 cm from the field edge by using the designed shielding device in most delivery techniques. In patients, the newly designed shielding structures can effectively eliminate up to about 49% of the foetal dose generated from various gantry angles used in VMAT or IMRT. CONCLUSIONS: We designed a foetal shielding device consisting of three parts to effectively reduce the dose delivered to the foetus, and evaluated the device with various treatment techniques for a pregnant patient with brain tumour. The foetal shielding device shielded the scattered/leaked radiation from the treatment machine, and also effectively reduced internal scattering from the treatment area in the patient.

Occupational Performance of Hearing-Impaired and Normal-Hearing Workers in Korea.

BACKGROUND AND OBJECTIVES: This study aimed to investigate the occupational performance of Korean workers with and without hearing loss and analyze the hearing-related difficulties in the working environment. SUBJECTS AND PURPOSE: The Amsterdam checklist for hearing and work was used for the analyses and the occupational environments of the Korean workers were investigated. Out of 129 total participants, 86 workers experienced severe to profound hearing loss and 43 had the normal hearing ability. The hearing-impaired workers were recruited from two leading vocational centers and normal-hearing workers were their colleagues. RESULTS: The hearing-impaired workers were found to take fewer sick leaves and exhibited higher rates of permanent job statuses compared to the normal-hearing workers. Workers with hearing loss rarely detected background sound; however, they could perceive reverberation more frequently. They felt more satisfied with their careers than the normal hearing workers as they received social support and needed to put their effort into hearing for most hearing activities. Furthermore, the effort in hearing increased with the increase in job demand, job control, social support, and career satisfaction. The working hours per week increased with the increase in age, education level, job demand, job control, and social support. Different trends were observed in 9 out of 12 variables while comparing the data from the present study with that obtained from the hearing-impaired workers of the Netherlands, indicating a large difference between countries. CONCLUSIONS: Although the hearing-impaired Korean workers operate diligently with good job positions, it is necessary to enhance their acoustic environment and provide them social support. Considering the cultural background of the hearing-impaired workers, the development of suitable vocational rehabilitation programs and specific questionnaires is strongly recommended worldwide.

A respiratory-guided 4D digital tomosynthesis.

The aim of this research was to introduce and evaluate a respiratory-guided slow gantry rotation 4D digital tomosynthesis (DTS). For each of ten volunteers, two breathing patterns were obtained for 3 min, one under free breathing conditions and the other with visual respiratory-guidance using an in-house developed respiratory monitoring system based on pressure sensing. Visual guidance was performed using a 4 s cycle sine wave with an amplitude corresponding to the average of end-inhalation peaks and end-exhalation valleys from the free-breathing pattern. The scan range was 40 degrees for each simulation, and the frame rate and gantry rotation speed were determined so that one projection per phase should be included. Both acquisition time and the number of total projections to be acquired (NPA) were calculated. Applying the obtained respiration pattern and the corresponding sequence, virtual projections were acquired under a typical geometry of Varian on-board imager for two virtual phantoms, modified Shepp-Logan (mSL) and extended cardiac-torso (XCAT). For the XCAT, two different orientations were considered, anterior-posterior (i.e. coronal) and left-right (i.e. sagittal). Projections were sorted to ten phases and image reconstruction was made using a modified filtered back-projection. Reconstructed images were compared with the planned breathing data (i.e. ideal situation) by structural similarity index (SSIM) and normalized root-mean-square error (NRMSE). For each case, simulation with guidance (SwG) showed motion-related artefact reduction compared to that under free-breathing (SuFB). SwG required less NPA but provided slightly higher SSIM and lower NRMSE values in all phantom images than SuFB did. In addition, the distribution of projections per phase was more regular in SwG. Through the proposed respiratory-guided 4D DTS, it is possible to reduce imaging dose while improving image quality. (Institutional Review Board approval: MC17DESI0086).

Analysis of Output Levels of an MP3 Player: Effects of Earphone Type, Music Genre, and Listening Duration.

BACKGROUND AND OBJECTIVES: To prevent noise induced hearing losses caused by listening to music with personal listening devices for young adults, this study was aimed to measure output levels of an MP3 and to identify preferred listening levels (PLLs) depending on earphone types, music genres, and listening durations. SUBJECTS AND METHODS: Twenty-two normal hearing young adults (mean=18.82, standard deviation=0.57) participated. Each participant was asked to select his or her most PLLs when listened to Korean ballade or dance music with an earbud or an over-the-ear earphone for 30 or 60 minutes. One side of earphone was connected to the participant's better ear and the other side was connected to a sound level meter via a 2 or 6 cc-couplers. Depending on earphone types, music genres, and listening durations, loudness A-weighted equivalent (LAeq) and loudness maximum time-weighted with A-frequency sound levels in dBA were measured. RESULTS: Neither main nor interaction effects of the PLLs among the three factors were significant. Overall output levels of earbuds were about 10-12 dBA greater than those of over-the-ear earphones. The PLLs were 1.73 dBA greater for earbuds than over-the-ear earphones. The average PLL for ballad was higher than for dance music. The PLLs at LAeq for both music genres were the greatest at 0.5 kHz followed by 1, 0.25, 2, 4, 0.125, 8 kHz in the order. CONCLUSIONS: The PLLs were not different significantly when listening to Korean ballad or dance music as functions of earphone types, music genres, and listening durations. However, over-the-ear earphones seemed to be more suitable to prevent noise induce hearing loss when listening to music, showing lower PLLs, possibly due to isolation from the background noise by covering ears.

Acoustic Features and Cortical Auditory Evoked Potentials according to Emotional Statues of /u/, /a/, /i/ Vowels.

BACKGROUND AND OBJECTIVES: Although Ling 6 sounds are often used in the rehabilitation process, its acoustic features have not been fully analyzed and represented in cortical responses. Current study was aimed to analyze acoustic features according to gender and emotional statuses of core vowels of Ling 6 sounds, /u/, /a/, and /i/. Cortical auditory evoked potentials (CAEPs) were also observed in those vowels. SUBJECTS AND METHODS: Vowel sounds /u/, /a/, and /i/ out of Ling 6 sounds representing low, middle and high frequencies were recorded from normal 20 young adults. The participants watched relevant videos for 4-5 minutes in order for them to sympathize emotions with anger (A), happiness (H), and sadness (S) before producing vowels. And without any emotional salience, neutrally production was performed. The recording was extracted for 500 ms to select pure vowel portion of production. For analysis of CAEP, the latencies and amplitudes of P1, N1, P2, N2, N1-P2 were analyzed. RESULTS: Intensities of /u/, /a/, and /i/ were 61.47, 63.38, and 60.55 dB. The intensities of neutral (N), H, A, S were 60.60, 65.43, 64.21, and 55.75 dB for vowel /u/, vowel /a/ were 61.80, 68.98, 66.50, and 56.23 dB, and vowel /i/ were 59.34, 64.90, 61.90, and 56.05 dB. The statistical significances for vowel and emotion were found but not for gender. The fundamental frequency (F0) of vowels for N, A, H, and S were 168.04, 174.93, 182.72, and 149.76 Hz and the first formant were 743.75, 815.59, 823.32, and 667.62 Hz. The statistical significance of F0 was found by vowel, emotion, and gender. The latencies and amplitudes of CAEP components did not show any statistical significance according to vowel. CONCLUSIONS: Ling 6 sounds should be produced consistently in the rehabilitation process for considering their difference of intensities and frequencies according to speaker's emotions and gender. The vowels seemed to be interpreted as tonal stimuli for CAEP components of this study with similar acoustic features among them. Careful selection of materials is necessary to observe meaningful conclusion of CAEP measurement with vowel stimuli.

Assessing Computational Fractional Flow Reserve From Optical Coherence Tomography in Patients With Intermediate Coronary Stenosis in the Left Anterior Descending Artery.

BACKGROUND: Intravascular optical coherence tomography (OCT) imaging provides limited information on the functional assessment of coronary stenosis. We evaluated a new approach to OCT image-based computation modeling, which can be used to estimate the fractional flow reserve (FFR) in patients with intermediate coronary stenosis. METHODS AND RESULTS: Ninety-two patients with intermediate diameter stenosis in the left anterior descending artery underwent both FFR measurement with pressure wires and OCT examination. Using the OCT data, a computational fluid dynamics algorithm was used to calculate the computational FFR (FFROCT). The diagnostic performance of the FFROCT was assessed based on the pressure wire-based FFR. The median FFR and FFROCT values were 0.86 (0.79-0.89) and 0.89 (0.82-0.94), respectively. The average diameter stenosis in quantitative coronary angiography and area stenosis in OCT were 58.1±13.4% and 67.5±13.5%, respectively. The FFROCT was better correlated to the FFR than were the anatomic variables (r=0.72; P<0.001 versus r=0.46; P<0.001 for minimal luminal diameter on quantitative coronary angiography or r=0.57; P<0.001 for minimal lumen area on OCT). When functionally significant stenosis was defined as an FFR cutoff value of ≤0.8, FFROCT resulted in 88.0% accuracy, 68.7% sensitivity, and 95.6% specificity. The positive and negative predictive values were 84.2% and 89.0%, respectively. CONCLUSIONS: The computation of FFROCT enables assessment not only of anatomic information, but also of the functional significance of intermediate stenosis. This measurement may be a useful approach for the simultaneous evaluation of the functional and anatomic severity of coronary stenosis.

Reducing radiation dose by application of optimized low-energy x-ray filters to K-edge imaging with a photon counting detector.

K-edge imaging with photon counting x-ray detectors (PCXDs) can improve image quality compared with conventional energy integrating detectors. However, low-energy x-ray photons below the K-edge absorption energy of a target material do not contribute to image formation in the K-edge imaging and are likely to be completely absorbed by an object. In this study, we applied x-ray filters to the K-edge imaging with a PCXD based on cadmium zinc telluride for reducing radiation dose induced by low-energy x-ray photons. We used aluminum (Al) filters with different thicknesses as the low-energy x-ray filters and implemented the iodine K-edge imaging with an energy bin of 34-48 keV at the tube voltages of 50, 70 and 90 kVp. The effects of the low-energy x-ray filters on the K-edge imaging were investigated with respect to signal-difference-to-noise ratio (SDNR), entrance surface air kerma (ESAK) and figure of merit (FOM). The highest value of SDNR was observed in the K-edge imaging with a 2 mm Al filter, and the SDNR decreased as a function of the filter thicknesses. Compared to the K-edge imaging with a 2 mm Al filter, the ESAK was reduced by 66%, 48% and 39% in the K-edge imaging with a 12 mm Al filter for 50 kVp, 70 kVp and 90 kVp, respectively. The FOM values, which took into account the ESAK and SDNR, were maximized for 8, 6 to 8 and 4 mm Al filters at 50 kVp, 70 kVp and 90 kVp, respectively. We concluded that the use of an optimal low-energy filter thickness, which was determined by maximizing the FOM, could significantly reduce radiation dose while maintaining image quality in the K-edge imaging with the PCXD.

Experimental Study of an Easily Controlled Ultra-High-Resolution Pixel-Matched Parallel-Hole Collimator with a Small Cadmium Zinc Telluride Pixelated Gamma Camera System.

The aim of this study was to develop an easily controlled, ultra-high-resolution, tungsten parallel-hole collimator based on a small pixelated gamma camera system. A small cadmium zinc telluride (CZT) pixelated semiconductor detector (eValuator-2500 detector [eV product, Saxonburg, PA]) was evaluated. This detector is composed of an array of 51.2 × 0.8 × 3-mm3 individual CZT crystal elements. The ultra-high-resolution, pixel-matched, parallel-hole collimators consisted of six layers, with the same between the hole and pixel size. The basic characteristics of the imaging system, such as sensitivity and spatial resolution, was measured using a 57Co point source. The measured averages of sensitivity and spatial resolution varied depending on the septal heights of the ultra-high-resolution parallel-hole collimator and source-to-collimator distances. When the 30-mm septal height was at 1-cm source-to-collimator distance, the spatial resolution was approximately 0.85 mm. Using 5-mm septal height, over 0.3 cps/kBq sensitivity was achieved. One advantage of our system is the use of stacked collimators that can select the best combination of system sensitivity and spatial resolution. Our results demonstrated that the developed CZT-pixelated gamma camera system using an ultra-high-resolution parallel-hole collimator of various collimator geometric designs has potential as an effective instrument.

Quantitative material decomposition using spectral computed tomography with an energy-resolved photon-counting detector.

Dual-energy computed tomography (CT) techniques have been used to decompose materials and characterize tissues according to their physical and chemical compositions. However, these techniques are hampered by the limitations of conventional x-ray detectors operated in charge integrating mode. Energy-resolved photon-counting detectors provide spectral information from polychromatic x-rays using multiple energy thresholds. These detectors allow simultaneous acquisition of data in different energy ranges without spectral overlap, resulting in more efficient material decomposition and quantification for dual-energy CT. In this study, a pre-reconstruction dual-energy CT technique based on volume conservation was proposed for three-material decomposition. The technique was combined with iterative reconstruction algorithms by using a ray-driven projector in order to improve the quality of decomposition images and reduce radiation dose. A spectral CT system equipped with a CZT-based photon-counting detector was used to implement the proposed dual-energy CT technique. We obtained dual-energy images of calibration and three-material phantoms consisting of low atomic number materials from the optimal energy bins determined by Monte Carlo simulations. The material decomposition process was accomplished by both the proposed and post-reconstruction dual-energy CT techniques. Linear regression and normalized root-mean-square error (NRMSE) analyses were performed to evaluate the quantitative accuracy of decomposition images. The calibration accuracy of the proposed dual-energy CT technique was higher than that of the post-reconstruction dual-energy CT technique, with fitted slopes of 0.97-1.01 and NRMSEs of 0.20-4.50% for all basis materials. In the three-material phantom study, the proposed dual-energy CT technique decreased the NRMSEs of measured volume fractions by factors of 0.17-0.28 compared to the post-reconstruction dual-energy CT technique. It was concluded that the proposed dual-energy CT technique can potentially be used to decompose mixtures into basis materials and characterize tissues according to their composition.

Variable aperture controlled by microelectrofluidic iris.

This Letter presents an adaptive liquid iris based on microelectrofluidic technology with experimental results. In the microelectrofluidic iris (MEFI), the electrostatic force generated by electrowetting in a surface channel unbalances the Laplace pressure acting on two fluidic interfaces between air and a light-absorbing liquid in two connected surface channels in a chamber. Then, the changed net pressure makes the iris aperture of the liquid diaphragm adjustable. The present MEFI was designed to have a tunable range from 4.2 to 0.85 mm in diameter and a tuning ratio of 80%. The MEFI was fabricated with a transparent electrode patterned on three glass plates and two channel spacers. Concerning the optical and interfacial properties of the MEFI for its operation, an aqueous near-infrared dye used in optical coherence tomography (OCT) was forced into a ring shape as the driving liquid in the hydrophobic chamber. By switching the segmented concentric control electrodes in steps, digital operation of the MEFI was successfully observed with clear aperture stops. The measured turnaround speed was 80 mm/s, which is significantly higher than that for other comparable adaptive liquid irises. Due to a scalable aperture range with fast response, the concept of MEFI is expected to be widely applied in various optical systems that require high-quality imaging, as well as in real-time diagnostic OCT.

Adaptive optical probe design for optical coherence tomography and microscopy using tunable optics.

We present a tunable, adaptive optical imaging probe for multimodal imaging such as optical coherence tomography and microscopy. The probe is compatible with forward-looking scanning laser imaging devices such as an endoscope. The lens configuration includes a tunable iris and two varifocal lenses, both driven by microelectrofluidics, as well as several conventional fixed focus lenses. The modulation transfer function and spot size in the focal plane is evaluated, and we show using optical simulations that there are three possible imaging modes with different transverse resolutions and focal depths.

Zoom lens design using liquid lens for laparoscope.

Traditional laparoscopic optical systems consisting of about 30 lenses have low optical magnification. To magnify tissue during surgical operations, one must change from one laparoscope to another or use a magnifying adapter between the laparoscope and the sensor. Our work focuses on how to change the sag of a liquid lens while zooming from 1 × zoom, to 2 × , and 4 × in an optical design for a laparoscope. The design includes several lenses and two liquid lenses with variable focal lengths. A pair of laparoscopes for 3-D stereoscopy is placed within a tube 11 mm in diameter. The predicted depth resolution of tissue is 0.5 mm without interpolation at 4 × zoom.

Varifocal liquid lens based on microelectrofluidic technology.

This Letter presents a tunable liquid lens based on microelectrofluidic technology. In the microelectrofluidic lens (MEFL), electrowetting in the hydrophobic surface channel induces the Laplace pressure difference between two fluidic interfaces on the lens aperture and the surface channel. Then, the pressure difference makes the lens curvature tunable. In spite of the contact angle saturation, the narrow surface channel increases the Laplace pressure to have a wide range of optical power variation in the MEFL. The magnitude of the applied voltage determines the lens curvature in the analog mode MEFL. Digital operation is also possible when the control electrodes of the MEFL are patterned to have an array. The lens aperture and maximum surface channel diameter were designed to 3.2 mm and 6.4 mm, respectively, with a channel height of 0.2 mm for an optical power range between +210 and -30 D. By switching the control electrodes, the averaged transit time in steps and turnaround time were as low as 2.4 ms and 16.5 ms, respectively, in good agreement with the simulation results. It is expected that the proposed MEFL may be widely used with advantages of wide variation of the optical power with fast and precise controllability in a digital manner.

A Monte Carlo simulation study of the effect of energy windows in computed tomography images based on an energy-resolved photon counting detector.

The energy-resolved photon counting detector provides the spectral information that can be used to generate images. The novel imaging methods, including the K-edge imaging, projection-based energy weighting imaging and image-based energy weighting imaging, are based on the energy-resolved photon counting detector and can be realized by using various energy windows or energy bins. The location and width of the energy windows or energy bins are important because these techniques generate an image using the spectral information defined by the energy windows or energy bins. In this study, the reconstructed images acquired with K-edge imaging, projection-based energy weighting imaging and image-based energy weighting imaging were simulated using the Monte Carlo simulation. The effect of energy windows or energy bins was investigated with respect to the contrast, coefficient-of-variation (COV) and contrast-to-noise ratio (CNR). The three images were compared with respect to the CNR. We modeled the x-ray computed tomography system based on the CdTe energy-resolved photon counting detector and polymethylmethacrylate phantom, which have iodine, gadolinium and blood. To acquire K-edge images, the lower energy thresholds were fixed at K-edge absorption energy of iodine and gadolinium and the energy window widths were increased from 1 to 25 bins. The energy weighting factors optimized for iodine, gadolinium and blood were calculated from 5, 10, 15, 19 and 33 energy bins. We assigned the calculated energy weighting factors to the images acquired at each energy bin. In K-edge images, the contrast and COV decreased, when the energy window width was increased. The CNR increased as a function of the energy window width and decreased above the specific energy window width. When the number of energy bins was increased from 5 to 15, the contrast increased in the projection-based energy weighting images. There is a little difference in the contrast, when the number of energy bin is increased from 15 to 33. The COV of the background in the projection-based energy weighting images is only slightly changed as a function of the number of energy bins. In the image-based energy weighting images, when the number of energy bins were increased, the contrast and COV increased and decreased, respectively. The CNR increased as a function of the number of energy bins. It was concluded that the image quality is dependent on the energy window, and an appropriate choice of the energy window is important to improve the image quality.

Varifocal liquid-filled microlens operated by an electroactive polymer actuator.

We designed, fabricated, and characterized varifocal microlenses, whose focal length varies along with the deformation of a transparent elastomer membrane under hydraulic pressure tailored by electroactive polymer actuators. The microfluidic channel of the microlens was designed to be embedded between silicon and glass so that transient fluctuation of the optical fluid and elastomer membrane is effectively suppressed, and thus the microlens is optically stabilized in a reduced time. Multilayered poly(vinylidene fluoride-trifluoroethylene-clorotrifluoroethylene) actuators were also developed and integrated onto the microfluidic chambers. We demonstrated that the developed microlenses are suitable for use in microimaging systems to make their foci tunable.