Performance Investigation of Somfit Sleep Staging Algorithm.

Purpose: To investigate accuracy of the sleep staging algorithm in a new miniaturized home sleep monitoring device - Compumedics® Somfit. Somfit is attached to patient's forehead and combines channels specified for a pulse arterial tonometry (PAT)-based home sleep apnea testing (HSAT) device with the neurological signals. Somfit sleep staging deep learning algorithm is based on convolutional neural network architecture. Patients and Methods: One hundred and ten participants referred for sleep investigation with suspected or preexisting obstructive sleep apnea (OSA) in need of a review were enrolled into the study involving simultaneous recording of full overnight polysomnography (PSG) and Somfit data. The recordings were conducted at three centers in Australia. The reported statistics include standard measures of agreement between Somfit automatic hypnogram and consensus PSG hypnogram. Results: Overall percent agreement across five sleep stages (N1, N2, N3, REM, and wake) between Somfit automatic and consensus PSG hypnograms was 76.14 (SE: 0.79). The percent agreements between different pairs of sleep technologists' PSG hypnograms varied from 74.36 (1.93) to 85.50 (0.64), with interscorer agreement being greater for scorers from the same sleep laboratory. The estimate of kappa between Somfit and consensus PSG was 0.672 (0.002). Percent agreement for sleep/wake discrimination was 89.30 (0.37). The accuracy of Somfit sleep staging algorithm varied with increasing OSA severity - percent agreement was 79.67 (1.87) for the normal subjects, 77.38 (1.06) for mild OSA, 74.83 (1.79) for moderate OSA and 72.93 (1.68) for severe OSA. Conclusion: Agreement between Somfit and PSG hypnograms was non-inferior to PSG interscorer agreement for a number of scorers, thus confirming acceptability of electrode placement at the center of the forehead. The directions for algorithm improvement include additional arousal detection, integration of motion and oximetry signals and separate inference models for individual sleep stages.

Can Thermal Imaging Technique be Used to Predict the Healing Status of a Venous Leg Ulcer?

Venous leg ulcers (VLUs) are the most common chronic wound types in older populations, with many wounds not healing in the normal trajectory. Many older people with wounds are treated in their homes, currently assessed by monitoring the wound area over weeks to ascertain the potential for healing. A noncontact method using thermal imaging has been shown to predict the healing trajectory of diabetes-related foot ulcers, although has not been tested in VLU or the home setting. This project investigated the effectiveness of using thermal imaging to predict VLU healing in the homes of participants. Images of 78 ulcers were collected weekly using a thermal camera from 67 participants in their homes, at 5 consecutive time points. Final follow-up calls were undertaken at 12 weeks to ascertain healing status (healed/unhealed). Images were preprocessed and segmented and the area of the region of the wound was extracted. Kruskal-Wallis tests were performed to test the association of the change of areas over the 5 consecutive weeks with the healing status of the ulcers at 12 weeks. The 95% confidence interval plots were obtained to study the distribution of the area in the healed and unhealed cases. This study found that the difference in the imaged areas between unhealed ulcers at 12 weeks did not reach statistical significance using thermal imaging. Therefore, thermal images could not predict healing progression in VLUs when the images were taken in the homes of participants. Future research to improve the prediction of venous leg ulcer healing should include developing a protocol to standardize conditions, improve imaging process methods, and use machine learning.

Repeatability and Reproducibility of Retinal Fractal Dimension Measured with Swept-Source Optical Coherence Tomography Angiography in Healthy Eyes: A Proof-of-Concept Study.

The retinal vascular network fractal dimension (FD) could be a promising imaging biomarker. Our objective was to evaluate its repeatability and reproducibility in healthy eyes. A cross-sectional study was undertaken with young, healthy volunteers who had no reported cardiac risk factors or ocular disease history. For each participant, three SS-OCTA images (12 × 12 mm) were acquired using the Plex Elite 9000 (Carl Zeiss Meditec AG, Jena, Germany) by two ophthalmologists. Automated segmentation was obtained from both the superficial and deep capillary plexuses. FD was estimated by box counting. The intraclass correlation coefficients (ICC) were used as measures for repeatability and reproducibility. A total of 43 eyes of healthy volunteers were included. The mean ± standard deviation (SD) age was 30 ± 6.2 years. The results show good repeatability. The ICC was 0.722 (95% CI, 0.541-0.839) in the superficial capillary plexus and 0.828 (95% CI, 0.705-0.903) in the deep capillary plexus. For reproducibility, the ICC was 0.651 (95% CI, 0.439-0.795) and 0.363 (95% CI, 0.073-0.596) at the superficial and deep capillary plexus, respectively. In this study, the FD of the vascular network measured via SS-OCTA showed good repeatability and reproducibility in healthy participants.

Thermal imaging potential and limitations to predict healing of venous leg ulcers.

Area analysis of thermal images can detect delayed healing in diabetes foot ulcers, but not venous leg ulcers (VLU) assessed in the home environment. This study proposes using textural analysis of thermal images to predict the healing trajectory of venous leg ulcers assessed in home settings. Participants with VLU were followed over twelve weeks. Digital images, thermal images and planimetry of wound tracings of the ulcers of 60 older participants was recorded in their homes by nurses. Participants were labelled as healed or unhealed based on status of the wound at the 12th week follow up. The weekly change in textural features was computed and the first two principal components were obtained. 60 participants (aged 80.53 ± 11.94 years) with 72 wounds (mean area 21.32 ± 51.28cm2) were included in the study. The first PCA of the change in textural features in week 2 with respect to week 0 were statistically significant for differentiating between healed and unhealed cases. Textural analysis of thermal images is an effective method to predict in week 2 which venous leg ulcers will not heal by week 12 among older people whose wounds are being managed in their homes.

Association between Optical Coherence Tomography and Fluorescein Angiography based retinal features in the diagnosis of Macular Edema.

The association between optical coherence tomography (OCT) and the geometrical vascular parameters obtained from the fluorescein angiography (FA) of the eyes with macular edema (ME) was investigated. Data from 82 untreated eyes with ME were studied. Fractal dimension (FD), simple tortuosity, branching angle, total angle count and vessel to background ratio were the five vasculature parameters from FA that were studied. The four OCT features measured were central retinal/foveal thickness, average para-fovea thickness, average peri-fovea thickness and OCT volume. The four OCT parameters showed a significant difference between ME requiring treatment (MERT) and non-MERT eyes with the central retinal thickness (threshold at 300 µm) and average para-fovea thickness (threshold at 338.5 µm) as most significant. The results also indicate that FD from the FA of retinal vessels in the macular region was associated with the changes in retinal thickness and that OCT parameters can potentially be used for directly identifying ME.

Agreement study between color and IR retinal images based on retinal vasculature morphological parameters.

BACKGROUND: Color fundus photography have been extensively used to explore the link between retinal morphology changes associated with various disease i.e. Diabetic Retinopathy, Glaucoma. The development of multimodal imaging system that integrates Infrared Scanning Laser Ophthalmoscope (IR-SLO) and Optical Coherence Tomography (OCT) could help in studying these diseases at an early stage. The aim of this study was to test the agreement between the retinal vasculature parameters from the Infrared images obtained from optical coherence tomography and color fundus imaging. METHODS: The IR and Color retinal images were obtained from 16 volunteer participants and seven retinal vessel parameters, i.e. Fractal Dimension (FD), Average Angle (ABA), Total Angle Count (TAC), Tortuosity (ST), Vessel/Background ratio (VBR), Central Retinal Arteriolar Equivalent (CRAE) and Central Retinal Venular Equivalent (CRVE) were extracted from these retinal images using Retinal Image Vasculature Assessment software (RIVAS) and Integrative Vessel Analysis (IVAN). RESULTS: The Bland Altman plot was used to investigate the agreement between the two modalities. The paired sample t-test was used to assess the presence of fixed bias and the slope of Least Square Regression (LSR) line for the presence of proportional bias. The paired sample t-test showed that there was no statistically significant difference between Color and IR based on retinal vessel features (all p values > 0.05). LSR also revealed no statistically significant difference in the retinal vessel features between Color and IR. CONCLUSION: This study has revealed that there is a fair agreement between Color and IR images based on retinal vessel features. This research has shown that it is possible to use IR images of the retina to measure the retinal vasculature parameters which has the advantage of being flash-less, can be used even if there is opacity due to cataract, and can be performed along with OCT on the same device.

The association of temperature of Diabetic Foot Ulcers with Chronic Kidney Disorder.

The area of the diabetic foot ulcer (DFU) and its reduction over weeks is used for assessment in clinical practices; however, literature reports that this is not reliable parameter. This work has investigated the association of change in the mean temperature of the ulcers with three clinical conditions relevant to wound healing, i.e. peripheral vascular disease (PVD), chronic kidney disease (CKD) and ischemic heart disease (IHD). Thermal and RGB images of 23 DFUs of the first two weeks of ulceration were studied. One-way ANOVA was performed on the change in mean temperature of the ulcers and change in area and it was found that the weekly change in mean temperature was higher for patients with CKD (p-value=0.009). Also, change in area measured from RGB images did not show any association with the clinical conditions. The application of this work is that the temperature obtained from thermal image of the ulcer can be used as a prognostic parameter for its assessment.

Relation Between Retinal Vasculature and Retinal Thickness in Macular Edema.

This study has investigated the relationship of the retinal vasculature and the retinal thickness for Macular Edema (ME) subjects. Ninety sets of Fluorescein Angiograph (FA) and Optical Coherence Tomography (OCT) of 54 participants were analyzed. Multivariate analysis using the binary logistic regression model was used to study the association between vessel parameters and retinal thickness. The results reveal retinal vessel feature i.e. fractal dimension (FD) as the most sensitive parameter to the changes in retinal thickness associated with ME. Thus, indicating a direct relationship between the retinal vasculature and retinal thickness which is caused due to neovascular causing exudates, leakages and hemorrhages, with applications for alternate modality for detection of ME.

Exudate detection in fundus images using deeply-learnable features.

Presence of exudates on a retina is an early sign of diabetic retinopathy, and automatic detection of these can improve the diagnosis of the disease. Convolutional Neural Networks (CNNs) have been used for automatic exudate detection, but with poor performance. This study has investigated different deep learning techniques to maximize the sensitivity and specificity. We have compared multiple deep learning methods, and both supervised and unsupervised classifiers for improving the performance of automatic exudate detection, i.e., CNNs, pre-trained Residual Networks (ResNet-50) and Discriminative Restricted Boltzmann Machines. The experiments were conducted on two publicly available databases: (i) DIARETDB1 and (ii) e-Ophtha. The results show that ResNet-50 with Support Vector Machines outperformed other networks with an accuracy and sensitivity of 98% and 0.99, respectively. This shows that ResNet-50 can be used for the analysis of the fundus images to detect exudates.

Is Thermal Imaging a Useful Predictor of the Healing Status of Diabetes-Related Foot Ulcers? A Pilot Study.

INTRODUCTION: In clinical practice, both area and temperature of the ulcer have been shown to be effective in tracking the healing status of diabetes-related foot ulcer (DRFU). However, traditionally, the area of the DRFU is measured regardless of the temperature distribution. The current prospective, observational study used thermal imaging, as a more accurate tool, to measure both the area and the temperature of DRFU. We aimed to predict healing of DRFU using thermal imaging within the first 4 weeks of ulceration. METHOD: A pilot study was conducted where thermal and color images of 26 neuropathic DRFUs (11 healing vs 15 nonhealing) from individuals with type 1 or 2 diabetes were taken at the initial clinic visit (baseline), at week 2, and at week 4. The thermal images were segmented into isothermal patches to identify the wound boundary and area corresponding to temperature distribution. Five parameters were obtained: temperature of the wound bed, area of the isothermal patch of the wound bed, area of isothermal patch of periwound, number of isolated isothermal patches of the wound region, and physical wound bed area from color image. The ulcers were also measured by experienced podiatrists over 4 consecutive weeks and used as the healing reference. RESULTS: For healing cases, the ratio of the area of the wound bed to its baseline measured using thermal images was found to be significantly lower at 2 weeks compared to nonhealing cases and this corresponded with a 50% reduction in area of DRFU at 4 weeks (group rank-based nonparametric analysis of variance P = .036). In comparison, neither the planimetric area measured using color images nor the temperature of the wound bed was associated with the healing. CONCLUSION: This study of 26 patients demonstrates that change in the isothermal area of DRFU can predict the healing status at week 4. Thermal imaging of DRFUs has the advantage of incorporating both area and temperature allowing for early prediction of the healing of these ulcers. Further studies with greater sample sizes are required to test the significance of these results.

Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms.

BACKGROUND: Convolution neural networks have been considered for automatic analysis of fundus images to detect signs of diabetic retinopathy but suffer from low sensitivity. METHODS: This study has proposed an alternate method using probabilistic output from Convolution neural network to automatically and simultaneously detect exudates, hemorrhages and microaneurysms. The method was evaluated using two approaches: patch and image-based analysis of the fundus images on two public databases: DIARETDB1 and e-Ophtha. The novelty of the proposed method is that the images were analyzed using probability maps generated by score values of the softmax layer instead of the use of the binary output. RESULTS: The sensitivity of the proposed approach was 0.96, 0.84 and 0.85 for detection of exudates, hemorrhages and microaneurysms, respectively when considering patch-based analysis. The results show overall accuracy for DIARETDB1 was 97.3% and 86.6% for e-Ophtha. The error rate for image-based analysis was also significantly reduced when compared with other works. CONCLUSION: The proposed method provides the framework for convolution neural network-based analysis of fundus images to identify exudates, hemorrhages, and microaneurysms. It obtained accuracy and sensitivity which were significantly better than the reported studies and makes it suitable for automatic diabetic retinopathy signs detection.

High-Resolution Spectral Analysis Accurately Identifies the Bacterial Signature in Infected Chronic Foot Ulcers in People With Diabetes.

Diabetic foot infections are a major cause of hospitalization, and delayed treatment can lead to numerous complications. The aim of this research was to investigate high-resolution spectroscopy of the wound center and periwound area for real-time estimation of multispectral signature of bacteria at the base of diabetic foot ulcers. We investigated the spectrum of the reflected visual light from diabetic foot ulcers and developed a method that identifies the presence of bacteria in the wound infections. We undertook a prospective pilot study on 18 patients with type 1 and type 2 diabetes and chronic diabetic foot ulcers. The spectral coefficients were directly compared with the results from the wound swab. The results of the multispectral analysis demonstrated 100% sensitivity, with 100% negative predictive values of identifying the presence of the bacteria, which was the cause of the infection in the wound. The results of our study suggest that the changes in the multispectral properties of the wound can be used to identify the presence of bacteria in the infected area using a noninvasive device without any contact with the wound. This technique holds great promise for real-time objective evaluation of the wound infection status beyond the standard visual assessment of diabetic foot ulcers.

A novel method for segmentation of Infrared Scanning Laser Ophthalmoscope (IR-SLO) images of retina.

Retinal vessel segmentation forms an essential element of automatic retinal disease screening systems. The development of multimodal imaging system with IR-SLO and OCT could help in studying the early stages of retinal disease. The advantages of IR-SLO to examine the alterations in the structure of retina and direct correlation with OCT can be useful for assessment of various diseases. This paper presents an automatic method for segmentation of IR-SLO fundus images based on the combination of morphological filters and image enhancement techniques. As a first step, the retinal vessels are contrasted using morphological filters followed by background exclusion using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Bilateral filtering. The final segmentation is obtained by using Isodata technique. Our approach was tested on a set of 26 IR-SLO images and results were compared to two set of gold standard images. The performance of the proposed method was evaluated in terms of sensitivity, specificity and accuracy. The system has an average accuracy of 0.90 for both the sets.

A novel approach for quantification of contour irregularities of diabetic foot ulcers and its association with ischemic heart disease.

Diabetic Foot Ulcer (DFU) is a common and global health problem in patients suffering from Diabetes. Delayed healing of these ulcers lead to serious problems, such as infection and amputation, apart from the problems faced in carrying out daily activities. This work has hypothesized the change in irregularity of wound contour may be linked to patient's clinical conditions. So far, fractal dimension is known to describe the change in irregularity of the fractal objects with change in scale but it fails to show any relation with the medical conditions in case of DFUs. This work introduces a new approach to quantify the irregularity of the wound contour and studies its association with some medical conditions. It was found that the change in irregularity of the wound contour from week 1 to week 4 had a close relation with ischemic heart disease as an important factor affecting the healing of DFUs.

Personality and Total Health Through Life Project Eye Substudy: Methodology and Baseline Retinal Features.

PURPOSE: To describe the methodology and present the retinal grading findings of an older sample of australians with well-defined indices of neurocognitive function in the Personality and total Health (PATH) through life project. DESIGN: A cross-sectional study. METHODS: Three hundred twenty-six individuals from the PatH through life project were invited to participate. Participants completed a general questionnaire and 2-field, 45-degree nonmydriatic color digital retinal photography. Photographs were graded for retinal pathology according to established protocols. RESULTS: Two hundred fifty-four (77.9%) subjects, aged 72 to 78 years, agreed to participate in the eye substudy. gradable images of at least 1 eye were acquired in 211 of 254 subjects (83.1%). retinal photographic screening identified 1 or more signs of pathology in 130 of the 174 subjects (74.7%) with gradable images of both eyes. a total of 45 participants (17.7%) had self-reported diabetes and diabetic retinopathy was observed in 22 (48.9%) of these participants. CONCLUSIONS: This well-defined sample of older australians provides a unique opportunity to interrogate associations between retinal findings, including retinal vascular geometric parameters, and indices of neurocognitive function.

Automatic Analysis of Retinal Vascular Parameters for Detection of Diabetes in Indian Patients with No Retinopathy Sign.

This study has investigated the association between retinal vascular parameters with type II diabetes in Indian population with no observable diabetic retinopathy. It has introduced two new retinal vascular parameters: total number of branching angles (TBA) and average acute branching angles (ABA) as potential biomarkers of diabetes in an explanatory model. A total number of 180 retinal images (two (left and right) × two (ODC and MC) × 45 subjects (13 diabetics and 32 nondiabetics)) were analysed. Stepwise linear regression analysis was performed to model the association between type II diabetes with the best subset of explanatory variables (predictors), consisting of retinal vascular parameters and patients' demographic information. P value of the estimated coefficients (P < 0.001) indicated that, at α level of 0.05, the newly introduced retinal vascular parameters, that is, TBA and ABA together with CRAE, mean tortuosity, SD of branching angle, and VB, are related to type II diabetes when there is no observable sign of retinopathy.

Adaptive Higuchi's dimension-based retinal vessel diameter measurement.

This paper proposes the use of an adaptive model to measure the width of retinal vessels in fundus photographs. It is based on the hypothesis that the Higuchi's dimension of the cross-section is proportional to the vessel diameter. This approach does not require image segmentation and binarization of the vessels and therefore is suitable for machine based measurements. The model is developed using a synthetic image of the vessel of known diameter and added noise. As a first step, the vessel cross-section profiles were manually obtained and analyzed using Higuchi's fractal dimension method. A 3D model was formed using the relationship between Higuchi's dimension, vessel diameter and noise variance. This model was then evaluated using expert annotated REVIEW public database and solved for the vessel diameter for performance evaluation. The result showed a good agreement with state of the art techniques. The model is background noise tolerant, estimates the vessel width with subpixel precision and does not require manual intervention.

Zone specific fractal dimension of retinal images as predictor of stroke incidence.

Fractal dimensions (FDs) are frequently used for summarizing the complexity of retinal vascular. However, previous techniques on this topic were not zone specific. A new methodology to measure FD of a specific zone in retinal images has been developed and tested as a marker for stroke prediction. Higuchi's fractal dimension was measured in circumferential direction (FDC) with respect to optic disk (OD), in three concentric regions between OD boundary and 1.5 OD diameter from its margin. The significance of its association with future episode of stroke event was tested using the Blue Mountain Eye Study (BMES) database and compared against spectrum fractal dimension (SFD) and box-counting (BC) dimension. Kruskal-Wallis analysis revealed FDC as a better predictor of stroke (H = 5.80, P = 0.016, α = 0.05) compared with SFD (H = 0.51, P = 0.475, α = 0.05) and BC (H = 0.41, P = 0.520, α = 0.05) with overall lower median value for the cases compared to the control group. This work has shown that there is a significant association between zone specific FDC of eye fundus images with future episode of stroke while this difference is not significant when other FD methods are employed.

Relationship between diabetes and grayscale fractal dimensions of retinal vasculature in the Indian population.

BACKGROUND: Diabetes mellitus is rapidly increasing in the Indian population. The purpose of this study was to identify changes in the retinal vasculature of diabetic people, ahead of visual impairments. Grayscale Fractal Dimension (FD) analysis of retinal images was performed on people with type 2 diabetes from an Indian population. METHODS: A cross-sectional study comprising 189 Optic Disc (OD) centred retinal images of healthy and diabetic individuals aged 14 to 73 years was conducted. Grayscale Box Counting FD of these retinal photographs was measured without manual supervision. Statistical analysis was conducted to determine the difference in the FD between diabetic and healthy (non-diabetic) people. RESULTS: The results show that grayscale FD values for diabetic cases are higher compared to controls, irrespective of the gender. It was also observed that FD was higher for male compared with females. CONCLUSIONS: There is difference in the grayscale fractal dimension of retinal vasculature of diabetic patients and healthy subjects, even when there is no reported retinopathy.

Adaptive colour transformation of retinal images for stroke prediction.

Identifying lesions in the retinal vasculature using Retinal imaging is most often done on the green channel. However, the effect of colour and single channel analysis on feature extraction has not yet been studied. In this paper an adaptive colour transformation has been investigated and validated on retinal images associated with 10-year stroke prediction, using principle component analysis (PCA). Histogram analysis indicated that while each colour channel image had a uni-modal distribution, the second component of the PCA had a bimodal distribution, and showed significantly improved separation between the retinal vasculature and the background. The experiments showed that using adaptive colour transformation, the sensitivity and specificity were both higher (AUC 0.73) compared with when single green channel was used (AUC 0.63) for the same database and image features.