The Purpose of the study was to develop a deep residual learning algorithm to screen for glaucoma from fundus photography and measure its diagnostic performance compared to Residents in ...Ophthalmology. A training dataset consisted of 1,364 color fundus photographs with glaucomatous indications and 1,768 color fundus photographs without glaucomatous features. A testing dataset consisted of 60 eyes of 60 glaucoma patients and 50 eyes of 50 normal subjects. Using the training dataset, a deep learning algorithm known as Deep Residual Learning for Image Recognition (ResNet) was developed to discriminate glaucoma, and its diagnostic accuracy was validated in the testing dataset, using the area under the receiver operating characteristic curve (AROC). The Deep Residual Learning for Image Recognition was constructed using the training dataset and validated using the testing dataset. The presence of glaucoma in the testing dataset was also confirmed by three Residents in Ophthalmology. The deep learning algorithm achieved significantly higher diagnostic performance compared to Residents in Ophthalmology; with ResNet, the AROC from all testing data was 96.5 (95% confidence interval CI: 93.5 to 99.6)% while the AROCs obtained by the three Residents were between 72.6% and 91.2%.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
We sought to construct and evaluate a deep learning (DL) model to diagnose early glaucoma from spectral-domain optical coherence tomography (OCT) images.
Artificial intelligence diagnostic tool ...development, evaluation, and comparison.
This multi-institution study included pretraining data of 4316 OCT images (RS3000) from 1371 eyes with open angle glaucoma (OAG) regardless of the stage of glaucoma and 193 normal eyes. Training data included OCT-1000/2000 images from 94 eyes of 94 patients with early OAG (mean deviation > −5.0 dB) and 84 eyes of 84 normal subjects. Testing data included OCT-1000/2000 from 114 eyes of 114 patients with early OAG (mean deviation > −5.0 dB) and 82 eyes of 82 normal subjects. A DL (convolutional neural network) classifier was trained using a pretraining dataset, followed by a second round of training using an independent training dataset. The DL model input features were the 8 × 8 grid macular retinal nerve fiber layer thickness and ganglion cell complex layer thickness from spectral-domain OCT. Diagnostic accuracy was investigated in the testing dataset. For comparison, diagnostic accuracy was also evaluated using the random forests and support vector machine models. The primary outcome measure was the area under the receiver operating characteristic curve (AROC).
The AROC with the DL model was 93.7%. The AROC significantly decreased to between 76.6% and 78.8% without the pretraining process. Significantly smaller AROCs were obtained with random forests and support vector machine models (82.0% and 67.4%, respectively).
A DL model for glaucoma using spectral-domain OCT offers a substantive increase in diagnostic performance.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP
The purpose of the current study was to evaluate the test-retest reproducibility and structure-function relationship of the MP-3 microperimeter, compared against the Humphrey Field Analyzer (HFA).
...Design: Reliability and validity study. Setting: Institutional, or clinical practice. Study Population: Thirty eyes of 30 primary open-angle glaucoma patients were enrolled. Observation Procedures: Visual fields (VF) were measured twice with the MP-3 and HFA instruments, using the 10-2 test grid pattern in both perimeters. Ganglion cell complex (GCC) thickness was measured using optical coherence tomography (OCT). Test-retest reproducibility was assessed using the mean absolute deviation (MAD) measure at all 68 VF test points, and also the intraclass correlation coefficient (ICC) of the repeated VF sensitivities. The structure-function relationship between VF sensitivities (measured with MP-3 or HFA) and GCC thickness (adjusted for the retinal ganglion cell displacement) was analyzed using linear mixed modeling. Main Outcome Measure: Reproducibility and structure-function relationship.
The average measurement duration with the HFA 10-2 was 7 minutes and 6 seconds (7m06s) ± 0m49s (mean ± standard deviation). A significantly (P < .001, paired Wilcoxon test) longer measurement duration was observed for the MP-3 test: 10m29s ± 2m55s. There were no significant differences in MAD and ICC values between HFA (MAD; 0.83 ± 0.69 dB and ICC: 0.89 ± 0.69, mean ± standard deviation) and MP-3 (MAD: 0.65 ± 0.67 dB and ICC: 0.89 ± 0.69). MP-3 VF sensitivities had a stronger structure-function relationship with GCC thickness compared to HFA.
The MP-3 microperimeter has a similar test-retest reproducibility to the HFA but a better structure-function relationship.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP
Objective: The present study proposes a new epileptic seizure prediction method through integrating heart rate variability (HRV) analysis and an anomaly monitoring technique. Methods: Because ...excessive neuronal activities in the preictal period of epilepsy affect the autonomic nervous systems and autonomic nervous function affects HRV, it is assumed that a seizure can be predicted through monitoring HRV. In the proposed method, eight HRV features are monitored for predicting seizures by using multivariate statistical process control, which is a well-known anomaly monitoring method. Results: We applied the proposed method to the clinical data collected from 14 patients. In the collected data, 8 patients had a total of 11 awakening preictal episodes and the total length of interictal episodes was about 57 h. The application results of the proposed method demonstrated that seizures in ten out of eleven awakening preictal episodes could be predicted prior to the seizure onset, that is, its sensitivity was 91%, and its false positive rate was about 0.7 times per hour. Conclusion: This study proposed a new HRV-based epileptic seizure prediction method, and the possibility of realizing an HRV-based epileptic seizure prediction system was shown. Significance: The proposed method can be used in daily life, because the heart rate can be measured easily by using a wearable sensor.
Abstract
Perovskite CH
3
NH
3
PbI
3
exhibits outstanding photovoltaic performances, but the understanding of the atomic motions remains inadequate even though they take a fundamental role in ...transport properties. Here, we present a complete atomic dynamic picture consisting of molecular jumping rotational modes and phonons, which is established by carrying out high-resolution time-of-flight quasi-elastic and inelastic neutron scattering measurements in a wide energy window ranging from 0.0036 to 54 meV on a large single crystal sample, respectively. The ultrafast orientational disorder of molecular dipoles, activated at ∼165 K, acts as an additional scattering source for optical phonons as well as for charge carriers. It is revealed that acoustic phonons dominate the thermal transport, rather than optical phonons due to sub-picosecond lifetimes. These microscopic insights provide a solid standing point, on which perovskite solar cells can be understood more accurately and their performances are perhaps further optimized.
We previously reported that the visual field (VF) prediction model using the variational Bayes linear regression (VBLR) is useful for accurately predicting VF progression in glaucoma (Invest ...Ophthalmol Vis Sci. 2014, 2018). We constructed a VF measurement algorithm using VBLR, and the purpose of this study was to investigate its usefulness.
122 eyes of 73 patients with open-angle glaucoma were included in the current study. VF measurement was performed using the currently proposed VBLR programme with AP-7700 perimetry (KOWA). VF measurements were also conducted using the Swedish interactive thresholding algorithm (SITA) standard programme with Humphrey field analyser. VF measurements were performed using the 24-2 test grid. Visual sensitivities, test-retest reproducibility and measurement duration were compared between the two algorithms.
Mean mean deviation (MD) values with SITA standard were -7.9 and -8.7 dB (first and second measurements), whereas those with VBLR-VF were -8.2 and -8.0 dB, respectively. There were no significant differences across these values. The correlation coefficient of MD values between the 2 algorithms was 0.97 or 0.98. Test-retest reproducibility did not differ between the two algorithms. Mean measurement duration with SITA standard was 6 min and 02 s or 6 min and 00 s (first or second measurement), whereas a significantly shorter duration was associated with VBLR-VF (5 min and 23 s or 5 min and 30 s).
VBLR-VF reduced test duration while maintaining the same accuracy as the SITA-standard.
Glassy magnetic behavior has been observed in a wide range of crystalline magnetic materials called spin glass. Here, we report spin glass behavior in a structural glass of a magnetic ionic liquid, ...C4mimFeCl
. Magnetization measurements demonstrate that an antiferromagnetic ordering occurs at T
= 2.3 K in the crystalline state, while a spin glass transition occurs at T
= 0.4 K in the structural glass state. In addition, localized magnetic excitations were found in the spin glass state by inelastic neutron scattering, in contrast to spin-wave excitations in the ordered phase of the crystalline sample. The localized excitation was scaled by the Bose population factor below T
and gradually disappeared above T
. This feature is highly reminiscent of boson peaks commonly observed in structural glasses. We suggest the "magnetic" boson peak to be one of the inherent dynamics of a spin glass state.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
To investigate the effects of cataract surgery on corneal biomechanics and intraocular pressure (IOP) measured with the updated Corvis ST tonometer (CST).
Prospective, interventional case series ...study.
This study included 39 eyes of 39 cataract patients. CST measurements were performed at presurgery (Pre) as well as 1 week (1W), 1 month (1M), and 3 months (3M) postsurgery. The following CST parameters were recorded: deformation amplitude max (DA max), DA ratio max 1 mm and 2 mm, integrated radius, stiffness parameter at applanation 1 (SP A1), Ambrosio relational thickness to the horizontal profile (ARTh), Corvis biomechanical index (CBI), central corneal thickness (CCT), noncorrected intraocular pressure (IOPnct), and biomechanically corrected IOP (bIOP). IOP was also measured with Goldmann applanation tonometry and the noncontact tonometer CT-90A. All measurements were compared at each period using the linear mixed model, with and without adjustment for bIOP and CCT.
All IOP measurements decreased over time (P < .01). CCT was increased at 1W and 3M (P < .01), whereas ARTh was decreased at 1W and 1M (P < .01), but returned to its Pre level at 3M. DA max and Integrated radius were increased at 3M (P < .01), whereas SP A1 was decreased at 3M (P < .01). CBI was increased at 1W (P < .01), but returned to its Pre level at 1M.
IOP and Corneal biomechanical properties are changed after cataract surgery. In particular, SP A1 decreases while DA max and integrated radius increase, even at 3M, suggesting a less stiff cornea.
•Corvis ST measured changes in corneal biomechanical properties after cataract surgery.•Stiffness parameter at applanation 1 decreases while maximum deformation amplitude and integrated radius increase after cataract surgery.•These changes suggest a less stiff cornea after cataract surgery.•Goldmann applanation tonometry–measured intraocular pressure was deceased.•Corvis ST's biomechanically corrected intraocular pressure was also decreased.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP
The purpose of this study was to investigate the association between visual acuity (VA) and reproducibility of test-retest visual field (VF) measurements in glaucoma patients.
Subjects were comprised ...of 627 eyes of 627 open-angle glaucoma patients. The reproducibility of two Humphrey VFs (24-2 or 30-2 Swedish Interactive Threshold Algorithm tests) examined twice within the period of 3 months was calculated using the root mean squared error (RMSE) of each VF test point's sensitivity. Visual acuity was measured once at the time of either of the VF measurements. Linear modeling was used to investigate the relationship between reproducibility of VF tests (RMSE) and the following variables: mean total deviation value (mTD), fixation losses (FLs), false positives (FPs), false negatives (FNs), refractive error, age, and VA.
The optimal model to predict test-retest variability (RMSE) of VFs included age, VA, mTD, and FNs as dependent variables. Root mean squared error was significantly larger in eyes with logMAR VA > 0.5 than in eyes with logMAR VA ≤ 0.
Reproducibility of VF tests becomes poor with the deterioration of VA. Careful consideration is needed when a patient's logMAR VA exceeds 0.5.
To validate a deep learning algorithm to diagnose glaucoma from fundus photography obtained with a smartphone.
A training dataset consisting of 1364 colour fundus photographs with glaucomatous ...indications and 1768 colour fundus photographs without glaucomatous features was obtained using an ordinary fundus camera. The testing dataset consisted of 73 eyes of 73 patients with glaucoma and 89 eyes of 89 normative subjects. In the testing dataset, fundus photographs were acquired using an ordinary fundus camera and a smartphone. A deep learning algorithm was developed to diagnose glaucoma using a training dataset. The trained neural network was evaluated by prediction result of the diagnostic of glaucoma or normal over the test datasets, using images from both an ordinary fundus camera and a smartphone. Diagnostic accuracy was assessed using the area under the receiver operating characteristic curve (AROC).
The AROC with a fundus camera was 98.9% and 84.2% with a smartphone. When validated only in eyes with advanced glaucoma (mean deviation value < -12 dB, N=26), the AROC with a fundus camera was 99.3% and 90.0% with a smartphone. There were significant differences between these AROC values using different cameras.
The usefulness of a deep learning algorithm to automatically screen for glaucoma from smartphone-based fundus photographs was validated. The algorithm had a considerable high diagnostic ability, particularly in eyes with advanced glaucoma.
PDF
