Machine Learning Models for Predicting Facial Nerve Palsy in Parotid Gland Surgery for Benign Tumors

Alkan

Clin Psychopharmacol Neurosci

2022

Objective: The attention deficit hyperactivity disorder has a negative impact on the child's educational life and relationships with the social environment during childhood and adolescence. The connection between temperament traits and The attention deficit hyperactivity disorder has been proven by various studies. As far as we know, there is no machine learning study to diagnose. The attention deficit hyperactivity disorder in a dataset created using temperament characteristics. Methods: Machine learning-based semi-automatic/fully automatic expert decision support systems are frequently used for the diagnosis of various diseases. In this study, it was aimed to reveal the success of a semi-automatic expert decision support system in the diagnosis of attention deficit hyperactivity disorder by using temperament characteristics. The high classification success achieved is a resource for a potential diagnosis of attention deficit hyperactivity disorder expert decision support system. In this respect, this study includes original qualities and innovations. Results: Many different deep learning methods were used in the research. Deep learning methods are models that achieve high success by using a large number of images in various image processing competitions. The images of the signals in the data set were first obtained by Continuous Wavelet Transform. The highest classification success in our data set was obtained with the Squeeze Net model with 88.33%. Conclusion: The model we propose shows that an automatic system based on artificial intelligence can be created, as well as revealing the relationship between temperament characteristics in the diagnosis of attention deficit hyperactivity in the data set we created.

“…Figure 8 shows how to calculate the accuracy, sensitivity, specificity, precision, and negative predictive values used to measure the performance of machine learning methods [ 32 - 34 ].…”

Section: Methodsmentioning

confidence: 99%

Automatic Diagnosis of Attention Deficit Hyperactivity Disorder with Continuous Wavelet Transform and Convolutional Neural Network

Alkan

Clin Psychopharmacol Neurosci

2022

“…Figure 3 shows simply how the values of accuracy (accuracy), sensitivity, specificity, precision, and negative prediction value used to measure the success of machine learning methods are calculated 24 .…”

Section: Methodsmentioning

confidence: 99%

Application of deep learning and classical machine learning methods in the diagnosis of attention deficit hyperactivity disorder according to temperament features

Alkan

Concurrency and Computation

2022

Attention deficit hyperactivity disorder (ADHD) is a common childhood neurodevelopmental disorder with symptoms of attention deficit, hyperactivity and impulsivity, with a prevalence of 8%-12% worldwide. Various studies have revealed that there may be a relationship between ADHD and temperament traits in the etiology of ADHD, which has a multifactorial etiology. According to our knowledge, there is no study in the literature that determines the use of machine learning methods in diagnosing ADHD uses a data set created with temperament characteristics. Different methods were used in this first study. The study included 60 ADHD patients and 60 control group children. The test scores of these children were collected from the Department of Child and Adolescent Psychiatry, Kahramanmaraş Sütçü İmam University Medical Faculty, after obtaining the necessary ethics committee permission. ADHD diagnosis was made according to DSM-5 classification. According to temperament characteristics, the highest classification success in ADHD diagnosis was calculated as 92.5% in decision tree method. Long short term memory (LSTM), one of the deep learning methods, achieved 88% classification success. The success of both methods is quite high and they have been compared with some ADHD classification studies in the literature.

“…Also, a three-dimensional facial database was too small to draw reliable findings. Another work [17] proposed a new model in deep machine learning in parotid polyps' surgery to detect facial nerve palsy by using the k-nearest neighbor algorithm that achieved values specificity of more than 0.9 performance. This work used 345 patients from 356 patients with parotid gland tumors, containing 192 males and 153 females aged between 18 and 87.…”

Section: Introductionmentioning

confidence: 99%

Computer Vision System for Facial Palsy Detection

Amsalam

Al-Naji

Daeef

et al. 2023

Facial palsy (FP) is a disorder that affects the seventh facial nerve, which makes the patient unable to control facial movements and expressions with other vital activities. It affects one side of the face, and it is usually diagnosed by the asymmetry of the two sides of the face through visual inspection by a doctor. However, the visual inspection is human-based, which is prone to errors because the doctor is exposed to omission due to fatigue and work stress. Therefore, it is important to develop a new method for detecting FP through artificial intelligence and use a more accurate computerized system to reduce the effort and cost of patients and increase the accuracy of diagnosis. This work aims to establish a safe, useful and high-accuracy diagnostic system for FP that can be used by the patient and proposes to detect FP using a digital camera and deep learning techniques automatically. The system could be used by the patient himself at home without needing to visit the hospital. The proposed system trained 570 images, including 200 images of FP palsy. The proposed FP system achieved an accuracy of 98%. This confirms the effectiveness of the proposed system and makes it an efficient medical examination tool for detecting FP.