Three-dimensional deeply generated holography [Invited].

In this paper, we present a noniterative method for 3D computer-generated holography based on deep learning. A convolutional neural network is adapted for directly generating a hologram to reproduce a 3D intensity pattern in a given class. We experimentally demonstrated the proposed method with optical reproductions of multiple layers based on phase-only Fourier holography. Our method is noniterative, but it achieves a reproduction quality comparable with that of iterative methods for a given class.

Deep learning wavefront sensing.

We present a new class of wavefront sensors by extending their design space based on machine learning. This approach simplifies both the optical hardware and image processing in wavefront sensing. We experimentally demonstrated a variety of image-based wavefront sensing architectures that can directly estimate Zernike coefficients of aberrated wavefronts from a single intensity image by using a convolutional neural network. We also demonstrated that the proposed deep learning wavefront sensor can be trained to estimate wavefront aberrations stimulated by a point source and even extended sources.

QT PRODACT: a multi-site study of in vitro action potential assays on 21 compounds in isolated guinea-pig papillary muscles.

To construct a non-clinical database for drug-induced QT interval prolongation, the electrophysiological effects of 11 positive and 10 negative compounds on action potentials (AP) in guinea-pig papillary muscles were investigated in a multi-site study according to a standard protocol. Compounds with a selective inhibitory effect on the rapidly activated delayed rectifier potassium current (IKr) prolonged action potential duration at 90% repolarization (APD90) in a concentration-dependent manner, those showing Ca2+ current (ICa) inhibition shortened APD30, and those showing Na+ current (INa) inhibition decreased action potential amplitude (APA) and Vmax. Some of the mixed ion-channel blockers showed a bell-shaped concentration-response curve for APD90, probably due to their blockade of INa and/or ICa, sometimes leading to a false-negative result in the assay. In contrast, all positive compounds except for terfenadine and all negative compounds with IKr-blocking activity prolonged APD30-90 regardless of their INa- and/or ICa-blocking activities, suggesting that APD30-90 is a useful parameter for evaluating the IKr-blocking activity of test compounds. Furthermore, the assay is highly informative regarding the modulation of cardiac ion channels by test compounds. Therefore, when APD90 and APD30-90 are both measured, the action potential assay can be considered a useful method for assessing the risk of QT interval prolongation in humans in non-clinical safety pharmacology studies.

QT PRODACT: evaluation of the potential of compounds to cause QT interval prolongation by action potential assays using guinea-pig papillary muscles.

Certain compounds that prolong QT interval in humans have little or no effect on action-potential (AP) duration used traditionally, but they inhibit rapidly-activated-delayed-rectifier potassium currents (IKr) and/or human ether-a-go-go-related gene (hERG) currents. In this study using isolated guinea-pig papillary muscles, we investigated whether new parameters in AP assays can detect the inhibitory effects of various compounds on IKr and/or hERG currents with high sensitivity. The difference in AP duration between 60% and 30% repolarization, 90% and 60% repolarization, and 90% and 30% repolarization (APD30-60, APD60-90, and APD30-90, respectively) were calculated as the new parameters. All the 15 IKr and/or hERG current inhibitors that have been reported (9 compounds) or not reported (6 compounds) to inhibit calcium currents prolonged APD30-60, APD60-90, and/or APD30-90; and 8 of the 15 inhibitors prolonged APD30-60, APD60-90, and/or APD30-90 more potently than APD90. The APD30-60, APD60-90, and APD30-90 measurements revealed no difference in sensitivity when evaluating the effects of the IKr and/or hERG current inhibitors on the three parameters. On the other hand, compounds with little or no effect on hERG currents had no effect on APD30-60, APD60-90, or APD30-90. Therefore, it is concluded that in AP assays using isolated guinea-pig papillary muscles, APD30-60, APD60-90, and APD30-90 are useful indexes for evaluating the inhibitory effects of compounds including mixed ion-channel blockers on IKr and/or hERG currents.

Cardiovascular assessment of ER-118585, a selective phosphodiesterase 5 inhibitor.

The aim of this study was to assess the cardiovascular effects of a selective phosphodiesterase 5 inhibitor ER-118585, 4-[(3-chloro-4-methoxybenzyl)amino]-1-(2-hydroxy-7-azaspiro[3.5]non-7-yl)-6-phthalazinecarbonitrile monohydrochloride. The present results indicated that 1) ER-118585 significantly inhibited the human ether-a-go-go related gene (HERG) tail current at 10 nM and above with an IC(50) value of 40.7 nM in human embryonic kidney 293 cells transfected with HERG cDNA; 2) ER-118585 at 100 and 1000 nM significantly increased the action potential duration (APD) at 50% and 90% repolarization in isolated papillary muscles of guinea pig; and 3) intravenous infusion of ER-118585 at 10 microg/kg/min significantly prolonged the QT interval by 10.5+/-1.6% from 281+/-2 ms to 311+/-6 ms in six anesthetized dogs subjected to atrial pacing. In consideration of both the plasma concentration of ER-118585 (984+/-78 nM, n=3) and its protein binding fraction (99.0+/-0.1%, n=5), the free plasma concentration was estimated at 9.8+/-0.8 nM, which is consistent with the minimum concentration of HERG current inhibition. In conclusion, these evaluation methods demonstrated that ER-118585 could prolong the QT interval via APD prolongation, attributable to the inhibition of the HERG potassium current.

Study of retinal nerve fiber layer thickness within normal hemivisual field in primary open-angle glaucoma and normal-tension glaucoma.

PURPOSE: To determine if nerve fiber layer thickness (NFLT) in glaucoma patients decreases before the development of visual field loss, and if there is a difference in the thinning of NFLT between primary open-angle glaucoma (POAG) and normal-tension glaucoma (NTG) eyes. METHODS: Thirty patients (33 eyes) with POAG and 31 patients (31eyes) with NTG, who had visual field defects localized in either the upper or the lower hemifield verified by Humphrey Field Analyzer (HFA), were measured for NFLT by scanning laser polarimetry (Nerve Fiber Analyzer). Twenty-three normal subjects (23 eyes) matched in refraction and age with the glaucoma patients were recruited as the control group. The total deviation (TD) in each hemifield obtained by HFA and the 180 degrees NFLT of each corresponding hemifield was calculated. Relationships between the TD of the normal or abnormal visual hemifield and the NFLT of the corresponding hemifield were compared among the POAG, NTG, and control groups. RESULTS: The NFLT of the corresponding normal hemifield was decreased both in the POAG group and in the NTG group when compared with the corresponding measurements in control subjects. In POAG eyes, thinning of the NFLT in the corresponding normal hemifield was more remarkable if the TD of the abnormal hemifield was greater, but this tendency was not observed in NTG eyes. CONCLUSION: NFLT is already decreased when the visual field is still normal both in POAG eyes and in NTG eyes. However, the pattern of retinal nerve fiber layer damage in POAG may differ from that in NTG.