Smart AMD prognosis through cellphone: an innovative localized AI-based prediction system for anti-VEGF treatment prognosis in nonagenarians and centenarians.

BACKGROUND AND OBJECTIVE: Age-related macular degeneration (AMD) is one of the most common reasons for blindness in the world today. The most common treatment for wet AMD is the intravitreal injections for inhibiting vascular-endothelial-derived growth factor (VEGF). This treatment usually involves multiple injections and thus multiple clinic visits, which not only causes increased cost on national health services but also causes exposure to the hospital environment, which is sometimes high risk considering current COVID crisis. The treatment, in spite of the above concerns, is usually effective. However, in some cases, either the medicine fails to produce the anticipated favourable outcome, resulting in waste of time, medication, efforts, and above all, psychological distress to the patients. Hence, early predictability of anatomical as well as functional effectiveness of the treatment appears to be a very desirable capability to have. METHOD: A machine learning approach using adaptive neuro-fuzzy inference system (ANFIS) of two-sample prediction model has been presented that requires only the baseline measurements and changes in visual acuity (VA) as well as macular thickness (MAC) after four months of treatment to estimate the values of VA and MAC at 8 and 12 months. In contrast to most of the AI techniques, ANFIS approach has shown the capability of the algorithm to work with very small dataset as well, which makes it a perfect candidate for the presented solution. RESULTS: The presented model has shown to have a very high accuracy (> 92%) and works in near-real-time scenarios. It has been converted into a smart phone App, OphnosisAMD, for convenient usage. With this App, the clinician can visualize the progression of the patient for a specific treatment and can decide on continuing or changing the treatment accordingly. The complete AI engine developed with the ANFIS algorithm is localized to the phone through the App, implying that there is no need for internet or cloud connectivity for this App to function. This makes it ideal for remote usage, especially under the current COVID scenarios. CONCLUSIONS: With a smart AI-based App on their fingertips, the presented system provides ample opportunity to the doctors to make a better decision based on the estimated progression, if the same drug is continued with (good/fair prognosis) or alternate treatment should be sought (bad prognosis). From a functional point of view, a prediction algorithm is triggered through simple entry of the relevant parameters (baseline and 4 months only). No internet/cloud connectivity is needed since the algorithm and the trained network are fully embedded in the App locally. Hence, using the App in remote and/or non-connected isolated areas is possible, especially in the secluded patients during the COVID scenarios.

Predicting infarction growth rate II using ANFIS-based binary particle swarm optimization technique in ischemic stroke.

Ischemic stroke, a severe medical condition triggered by a blockage of blood flow to the brain, leads to cell death and serious health complications. One key challenge in this field is accurately predicting infarction growth - the progressive expansion of damaged brain tissue post-stroke. Recent advancements in artificial intelligence (AI) have improved this prediction, offering crucial insights into the progression dynamics of ischemic stroke. One such promising technique, the Adaptive Neuro-Fuzzy Inference System (ANFIS), has shown potential, but it faces the 'curse of dimensionality' and long training times as the number of features increased. This paper introduces an innovative, automatic method that combines Binary Particle Swarm Optimization (BPSO) with ANFIS architecture, achieves reduction in dimensionality by reducing the number of rules and training time. By analyzing the Pearson correlation coefficients and P-values, we selected clinically relevant features strongly correlated with the Infarction Growth Rate (IGR II), extracted after one CT scan. We compared our model's performance with conventional ANFIS and other machine learning techniques, including Support Vector Regressor (SVR), shallow Neural Networks, and Linear Regression. •Inputs: Real data about ischemic stroke represented by clinically relevant features.•Output: An innovative model for more accurate and efficient prediction of the second infarction growth after the first CT scan.•Results: The model achieved commendable statistical metrics, which include a Root Mean Square Error of 0.091, a Mean Squared Error of 0.0086, a Mean Absolute Error of 0.064, and a Cosine distance of 0.074.

iMIGS: An innovative AI based prediction system for selecting the best patient-specific glaucoma treatment.

The use of AI-based techniques in healthcare are becoming more and more common and more disease-specific. Glaucoma is a disorder in eye that causes damage to the optic nerve which can lead to permanent blindness. It is caused by the elevated pressure inside the eye due to the obstruction to the flow of the drainage fluid (aqueous humor). Most recent treatment options involve minimally invasive glaucoma surgery (MIGS) in which a stent is placed to improve drainage of aqueous humor from the eye. Each MIGS surgery has a different mechanism of action, and the relative efficacy and chance of success is dependent on multiple patient-specific factors. Hence the ophthalmologists are faced with the critical question; which method would be better for a specific patient, both in terms of glaucoma control but also taking into consideration patient quality of life? In this paper, an Adaptive Neuro-Fuzzy Inference System (ANFIS) has been developed in the form of a Treatment Advice prediction system that will offer the clinician a suggested MIGS treatment from the baseline clinical parameters. ANFIS was used with a real-world MIGS data set which was a retrospective case series of 372 patients who underwent either of the four MIGS procedures from July 2016 till May 2020 at a single center in the UK.•Inputs used: Clinical measurements of Age, Visual Acuity, Intraocular Pressure (IOP), and Visual Field, etc.•Output Classes: iStent, iStent and Endoscopic Cyclophotocoagulation (ICE2), PreserFlo MicroShunt (PMS) and XEN-45).•Results: The proposed ANFIS system was found to be 91% accurate with high Sensitivity (80%) and Specificity (90%).

Dataset of usage pattern and energy analysis of an Internet of Things-enabled ceiling fan.

Many electrical appliances have progressed from sheer prototypes to viable products by being automated with the help of sensors and Internet of Things (IoT). In this data driven century, there aren't many data-centric solutions for the effective use of residential and commercial ceiling fans. For the said reason, sensors were installed on a remote-controlled BLDC ceiling fan, and a large amount of user data with environmental indicators such as temperature and humidity, was collected. This data along with the fan speed was logged to a cloud server over Wi-Fi using a Wi-Fi enabled microcontroller. The raw data consists of timestamp, temperature, humidity, and fan speed. The data is logged depending on the change of any parameter rather than a specific interval. The logged data is then visualized on the cloud server to monitor the usage patterns of the appliance and its subsequent energy consumption. The dataset is comprised of the fan data from the bedroom, living room, and lounge obtained by the resident's consent. This data is useful for data scientists, environmentalists, fan manufacturers, architects, social scientists, and several other field enthusiasts. The data can be analyzed based on monthly average temperature and humidity energy consumed, operational time per day or month and monthly/weekly summary of usage. Furthermore, by applying Artificial Intelligence (AI) algorithms on such data, it is feasible to extract patterns that indicate the appliance usage and identify changes in the daily routine. Many machine learning techniques can be applied on the dataset to introduce intelligent control of the appliance for adaptable operation without compromising on the comfort level of the user.

Estimating Blood Glucose Levels Using Machine Learning Models with Non-Invasive Wearable Device Data.

In 2019 alone, Diabetes Mellitus impacted 463 million individuals worldwide. Blood glucose levels (BGL) are often monitored via invasive techniques as part of routine protocols. Recently, AI-based approaches have shown the ability to predict BGL using data acquired by non-invasive Wearable Devices (WDs), therefore improving diabetes monitoring and treatment. It is crucial to study the relationships between non-invasive WD features and markers of glycemic health. Therefore, this study aimed to investigate accuracy of linear and non-linear models in estimating BGL. A dataset containing digital metrics as well as diabetic status collected using traditional means was used. Data consisted of 13 participants data collected from WDs, these participants were divided in two groups young, and Adult Our experimental design included Data Collection, Feature Engineering, ML model selection/development, and reporting evaluation of metrics. The study showed that linear and non-linear models both have high accuracy in estimating BGL using WD data (RMSE range: 0.181 to 0.271, MAE range: 0.093 to 0.142). We provide further evidence of the feasibility of using commercially available WDs for the purpose of BGL estimation amongst diabetics when using Machine learning approaches.

Wearable AI Reveals the Impact of Intermittent Fasting on Stress Levels in School Children During Ramadan.

Intermittent fasting has been practiced for centuries across many cultures globally. Recently many studies have reported intermittent fasting for its lifestyle benefits, the major shift in eating habits and patterns is associated with several changes in hormones and circadian rhythms. Whether there are accompanying changes in stress levels is not widely reported especially in school children. The objective of this study is to examine the impact of intermittent fasting during Ramadan on stress levels in school children as measured using wearable artificial intelligence (AI). Twenty-nine school children (aged 13-17 years and 12M / 17F ratio) were given Fitbit devices and their stress, activity and sleep patterns analyzed 2 weeks before, 4 weeks during Ramadan fasting and 2 weeks after. This study revealed no statistically significant difference on stress scores during fasting, despite changes in stress levels being observed for 12 of the participants. Our study may imply intermittent fasting during Ramadan poses no direct risks in terms of stress, suggesting rather it may be linked to dietary habits, furthermore as stress score calculations are based on heart rate variability, this study implies fasting does not interfere the cardiac autonomic nervous system.

How does hemodynamics affect rupture tissue mechanics in abdominal aortic aneurysm: Focus on wall shear stress derived parameters, time-averaged wall shear stress, oscillatory shear index, endothelial cell activation potential, and relative residence time.

An abdominal aortic aneurysm (AAA) is a critical health condition with a risk of rupture, where the diameter of the aorta enlarges more than 50% of its normal diameter. The incidence rate of AAA has increased worldwide. Currently, about three out of every 100,000 people have aortic diseases. The diameter and geometry of AAAs influence the hemodynamic forces exerted on the arterial wall. Therefore, a reliable assessment of hemodynamics is crucial for predicting the rupture risk. Wall shear stress (WSS) is an important metric to define the level of the frictional force on the AAA wall. Excessive levels of WSS deteriorate the remodeling mechanism of the arteries and lead to abnormal conditions. At this point, WSS-related hemodynamic parameters, such as time-averaged WSS (TAWSS), oscillatory shear index (OSI), endothelial cell activation potential (ECAP), and relative residence time (RRT) provide important information to evaluate the shear environment on the AAA wall in detail. Calculation of these parameters is not straightforward and requires a physical understanding of what they represent. In addition, computational fluid dynamics (CFD) solvers do not readily calculate these parameters when hemodynamics is simulated. This review aims to explain the WSS-derived parameters focusing on how these represent different characteristics of disturbed hemodynamics. A representative case is presented for spatial and temporal formulation that would be useful for interested researchers for practical calculations. Finally, recent hemodynamics investigations relating WSS-related parameters with AAA rupture risk assessment are presented. This review will be useful to understand the physical representation of WSS-related parameters in cardiovascular flows and how they can be calculated practically for AAA investigations.

Cranial reconstruction after decompressive craniectomy: prediction of complications using fuzzy logic.

INTRODUCTION: Cranial reconstruction after decompressive craniectomy (DC) has been shown to be associated with a relatively high complication rate (16.4%-34%) compared with standard neurosurgical procedures (2%-5%). Most studies that have previously attempted to formulate a multivariate model for identifying factors predictive of postoperative complications of cranioplasty either were unsuccessful or yielded conflicting results. Therefore, fuzzy logic-based fuzzy inference system (FIS), which has proven to be a useful tool for risk prediction in medical and surgical conditions, was used in this study to identify predictors of complications of cranioplasty. METHODS: A retrospective chart review of all the patients who underwent DC followed by elective cranioplasty at Aga Khan University Hospital, during a 10-year period (2000-2010), was carried out to collect data on 24 carefully selected preoperative variables or inputs. The proposed FIS had 24 inputs, 3 outputs, and a set of 7 fuzzy-based rules. All inputs were assigned degrees of membership, and complications were further divided into "severe," "minor," and "least" output classes with each of them representing 2 membership functions: "less" and "more." For each set of inputs, a specific portion of the hypersurface was masked out. The centroid of this subsurface represented the defuzzified output corresponding to 1 percentage value for each output. The maximum of these outputs for each of the 3 output classes was selected to be the final output class. Each output class was compared to the actual outcome of patients, and positive predictive value, negative predictive value, sensitivity, and specificity of FIS for predicting complications were calculated. RESULTS: A total of 89 patients (mean [SD] age, 33.1 [15.0] y; male-to-female ratio, 3:1) were included in the study. The common postoperative complications included seizures (14.6%), cerebrospinal fluid leak (4.5%), neurologic deficits (3.4%), hydrocephalus (3.4%), superficial wound infection (3.4%), and osteomyelitis (2.2%). The FIS correctly identified all 7 patients who developed severe complications after cranioplasty (true positives) and all 82 patients who did not develop severe complications (true negatives). Thus, the FIS has a sensitivity and specificity of 100% in predicting severe complications. CONCLUSIONS: Our study shows that the procedure of cranioplasty is associated with a high complication rate and that FIS has a 100% sensitivity and specificity in predicting severe complications after cranioplasty. It will prove to be an invaluable tool for clinicians once the results are validated by a similar prospective study with a larger sample size.

ANFIS-Net for automatic detection of COVID-19.

Among the most leading causes of mortality across the globe are infectious diseases which have cost tremendous lives with the latest being coronavirus (COVID-19) that has become the most recent challenging issue. The extreme nature of this infectious virus and its ability to spread without control has made it mandatory to find an efficient auto-diagnosis system to assist the people who work in touch with the patients. As fuzzy logic is considered a powerful technique for modeling vagueness in medical practice, an Adaptive Neuro-Fuzzy Inference System (ANFIS) was proposed in this paper as a key rule for automatic COVID-19 detection from chest X-ray images based on the characteristics derived by texture analysis using gray level co-occurrence matrix (GLCM) technique. Unlike the proposed method, especially deep learning-based approaches, the proposed ANFIS-based method can work on small datasets. The results were promising performance accuracy, and compared with the other state-of-the-art techniques, the proposed method gives the same performance as the deep learning with complex architectures using many backbone.

Computational methods for automated analysis of corneal nerve images: Lessons learned from retinal fundus image analysis.

Corneal and retinal imaging provide a descriptive view of the nerve and vessel structure present inside the human eye, in a non-invasive manner. This helps in ocular, or other, disease identification and diagnosis. However, analyzing these images is a laborious task and requires expertise in the field. Therefore, there is a dire need for process automation. Although a large body of literature is available for automated analysis of retinal images, research on corneal nerve image analysis has lagged due to several reasons. In this article, we cover the recent research trends in automated analysis of corneal and retinal images, highlighting the requirements for automation of corneal nerve image analysis, and the possible reasons impeding its research progress. We also present a comparative analysis of segmentation algorithms versus their processing power derived from the studies included in the survey. Due to the advancement in retinal image analysis and the implicit similarities in retinal and corneal images, we extract lessons from the former and suggest ways to apply them to the latter. This is presented as future research directions for automatic detection of neuropathy using corneal nerve images. We believe that this article will be extremely informative for computer scientists and medical professionals alike, as the former would be informed regarding the different research problems waiting to be addressed in the field, while the latter would be enlightened to what is required from them so as to facilitate computer scientists in their path towards finding effective solutions to the problems.

The role of EEG and EMG combined virtual reality gaming system in facial palsy rehabilitation - A case report.

BACKGROUND: The recovery rates for facial palsy are usually excellent; however, regularly patients present with problems with their fine facial movements that affect their emotional expressions. OBJECTIVE: To discover the viability and ease of using an Electroencephalogram (EEG) and Electromyography (EMG) combined Virtual Reality (VR) gaming system - the 'Oculus Rift' device in the evaluation and rehabilitation of facial palsy. DESIGN: Single case study. PATIENT INFORMATION: A young 23-year-old female with facial palsy. CLINICAL FINDINGS: Most of the patient's facial features were re-established within the recovery time frame, except for her right forehead and eyebrow movements. INTERVENTION: A 10 day exercise program (Day 2-11) with an immersive virtual reality device, which randomly shoots virtually animated white balls in an unpredictable and testing pattern. OUTCOME MEASURES: EEG and EMG patterns corresponding to the facial upper quadrant were taken at baseline, post-intervention, and at follow up. RESULTS: EMG and EEG investigation revealed a progressive improvement in the muscle activation in response to the impulsive and unpredictable activities in the virtual environment provided through the immersive VR device. CONCLUSION: The case report found a positive relationship between VR, facial upper quadrant EMG activation and EEG pattern changes following the intervention.

Infrared image enhancement using H(infinity) bounds for surveillance applications.

In this paper, two algorithms have been presented to enhance the infrared (IR) images. Using the autoregressive moving average model structure and H(infinity) optimal bounds, the image pixels are mapped from the IR pixel space into normal optical image space, thus enhancing the IR image for improved visual quality. Although H(infinity)-based system identification algorithms are very common now, they are not quite suitable for real-time applications owing to their complexity. However, many variants of such algorithms are possible that can overcome this constraint. Two such algorithms have been developed and implemented in this paper. Theoretical and algorithmic results show remarkable enhancement in the acquired images. This will help in enhancing the visual quality of IR images for surveillance applications.

Prediction of infarction volume and infarction growth rate in acute ischemic stroke.

The prediction of infarction volume after stroke onset depends on the shape of the growth dynamics of the infarction. To understand growth patterns that predict lesion volume changes, we studied currently available models described in literature and compared the models with Adaptive Neuro-Fuzzy Inference System [ANFIS], a method previously unused in the prediction of infarction growth and infarction volume (IV). We included 67 patients with malignant middle cerebral artery [MMCA] stroke who underwent decompressive hemicraniectomy. All patients had at least three cranial CT scans prior to the surgery. The rate of growth and volume of infarction measured on the third CT was predicted with ANFIS without statistically significant difference compared to the ground truth [P = 0.489]. This was not possible with linear, logarithmic or exponential methods. ANFIS was able to predict infarction volume [IV3] over a wide range of volume [163.7-600 cm3] and time [22-110 hours]. The cross correlation [CRR] indicated similarity between the ANFIS-predicted IV3 and original data of 82% for ANFIS, followed by logarithmic 70%, exponential 63% and linear 48% respectively. Our study shows that ANFIS is superior to previously defined methods in the prediction of infarction growth rate (IGR) with reasonable accuracy, over wide time and volume range.

Glandular Morphometrics for Objective Grading of Colorectal Adenocarcinoma Histology Images.

Determining the grade of colon cancer from tissue slides is a routine part of the pathological analysis. In the case of colorectal adenocarcinoma (CRA), grading is partly determined by morphology and degree of formation of glandular structures. Achieving consistency between pathologists is difficult due to the subjective nature of grading assessment. An objective grading using computer algorithms will be more consistent, and will be able to analyse images in more detail. In this paper, we measure the shape of glands with a novel metric that we call the Best Alignment Metric (BAM). We show a strong correlation between a novel measure of glandular shape and grade of the tumour. We used shape specific parameters to perform a two-class classification of images into normal or cancerous tissue and a three-class classification into normal, low grade cancer, and high grade cancer. The task of detecting gland boundaries, which is a prerequisite of shape-based analysis, was carried out using a deep convolutional neural network designed for segmentation of glandular structures. A support vector machine (SVM) classifier was trained using shape features derived from BAM. Through cross-validation, we achieved an accuracy of 97% for the two-class and 91% for three-class classification.

Fuzzy logic: A "simple" solution for complexities in neurosciences?

BACKGROUND: Fuzzy logic is a multi-valued logic which is similar to human thinking and interpretation. It has the potential of combining human heuristics into computer-assisted decision making, which is applicable to individual patients as it takes into account all the factors and complexities of individuals. Fuzzy logic has been applied in all disciplines of medicine in some form and recently its applicability in neurosciences has also gained momentum. METHODS: This review focuses on the use of this concept in various branches of neurosciences including basic neuroscience, neurology, neurosurgery, psychiatry and psychology. RESULTS: The applicability of fuzzy logic is not limited to research related to neuroanatomy, imaging nerve fibers and understanding neurophysiology, but it is also a sensitive and specific tool for interpretation of EEGs, EMGs and MRIs and an effective controller device in intensive care units. It has been used for risk stratification of stroke, diagnosis of different psychiatric illnesses and even planning neurosurgical procedures. CONCLUSIONS: In the future, fuzzy logic has the potential of becoming the basis of all clinical decision making and our understanding of neurosciences.

Autonomous corrosion detection in gas pipelines: a hybrid-fuzzy classifier approach using ultrasonic nondestructive evaluation protocols.

In this paper, a customized classifier is presented for the industry-practiced nondestructive evaluation (NDE) protocols using a hybrid-fuzzy inference system (FIS) to classify the corrosion and distinguish it from the geometric defects or normal/healthy state of the steel pipes used in the gas/petroleum industry. The presented system is hybrid in the sense that it utilizes both soft computing through fuzzy set theory, as well as conventional parametric modeling through H(infinity) optimization methods. Due to significant uncertainty in the power spectral density of the noise in ultrasonic NDE procedures, the use of optimal H(2) estimators for defect characterization is not so accurate. A more appropriate criterion is the H(infinity) norm of the estimation error spectrum which is based on minimization of the magnitude of this spectrum and hence produces more robust estimates. A hybrid feature set is developed in this work that corresponds to a) geometric features extracted directly from the raw ultrasonic A-scan data (which are the ultrasonic echo pulses in 1-Dtraveling inside the metal perpendicular to its 2 surfaces) and b) mapped features from the impulse response of the estimated model of the defect waveform under study. An experimental strategy is first outlined, through which the necessary data are collected as A-scans. Then, using the H(infinity) estimation approach, a parametric transfer function is obtained for each pulse. In this respect, each A-scan is treated as output from a defining function when a pure/healthy metal's A-scan is used as its input. Three defining states are considered in the paper; healthy, corroded, and defective, where the defective class represents metal with artificial or other defects. The necessary features are then calculated and are then supplied to the fuzzy inference system as input to be used in the classification. The resulting system has shown excellent corrosion classification with very low misclassification and false alarm rates.

Fuzzy Logic in neurosurgery: predicting poor outcomes after lumbar disk surgery in 501 consecutive patients.

BACKGROUND: Despite a lot of research into patient selection, a significant number of patients fail to benefit from surgery for symptomatic lumbar disk herniation. We have used Fuzzy Logic-based fuzzy inference system (FIS) for identifying patients unlikely to improve after disk surgery and explored FIS as a tool for surgical outcome prediction. METHODS: Data of 501 patients were retrospectively reviewed for 54 independent variables. Sixteen variables were short-listed based on heuristics and were further classified into memberships with degrees of membership within each. A set of 11 rules was formed, and the rule base used individual membership degrees and their values mapped from the membership functions to perform Boolean Logical inference for a particular set of inputs. For each rule, a decision bar was generated that, when combined with the other rules in a similar way, constituted a decision surface. The FIS decisions were then based on calculating the centroid for the resulting decision surfaces and thresholding of actual centroid values. The results of FIS were then compared with eventual postoperative patient outcomes based on clinical follow-ups at 6 months to evaluate FIS as a predictor of poor outcome. RESULTS: Fuzzy inference system has a sensitivity of 88% and specificity of 86% in the prediction of patients most likely to have poor outcome after lumbosacral miscrodiskectomy. The test thus has a positive predictive value of 0.36 and a negative predictive value of 0.98. CONCLUSION: Fuzzy inference system is a sensitive method of predicting patients who will fail to improve with surgical intervention.