Synthetic Data Generation System for AI-Based Diabetic Foot Diagnosis

Hyun, Jayun; Lee, Yongho; Son, Ha Min; Lee, Seo Hu; Pham, Vinh; Park, Ji Ung; Chung, Tai-Myoung

doi:10.1007/s42979-021-00667-9

Cited by 10 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Optimized Website Traffic Forecasting with Automatic Models and Optuna : A Study in Machine Learning dan Big Data Analytics diabetic foot sores. The model used multiplicative future regressors and demonstrated strong modeling power [18].These studies highlight the diverse applications and potential of Prophet and NeuralProphet in the field of time-series forecasting.…”

Section: Al Employed a Neuralprophet Model To Detectmentioning

confidence: 87%

Optimized Website Traffic Forecasting with Automatic Models and Optuna

Ahadi Ningrum,

Rifqi Mulyawan

2024

IIAI Letters on Informatics and Interdisciplinary Research

View full text Add to dashboard Cite

Although time-series forecasting has emerged as a key area of interest in recent years for analyzing historical data to make predictions about future trends, accurately forecasting web traffic can be challenging due to the dynamic nature of the internet and the many factors that can influence user behavior. Existing traffic flow prediction approaches primarily use simple models that are often inadequate for real-world applications. This research aims to develop an optimized machine learning model using FB-Prophet and NeuralProphet for forecasting website traffic and to compare their relative performance and effectiveness in predicting web traffic. This study aims to develop an optimized machine learning model using FB-Prophet and NeuralProphet for forecasting website traffic and to compare their performance in predicting web traffic. Our study found that both FB-Prophet and NeuralProphet models performed better than the simple models used in existing traffic flow prediction approaches. However, the NeuralProphet model outperformed FB-Prophet in terms of accuracy and computational efficiency. The best result obtained from the study was achieved by the NeuralProphet model, which had a Mean Average Error (MAE) of 25.61, Mean Square Error (MSE) of 1354, Root Mean Square Error (RMSE) of 5.060, and a Coefficient of Determination (R2) of 0.882, indicating its superior performance in accurately forecasting website traffic. The results suggest that an optimized machine learning model using NeuralProphet can be an effective way to forecast website traffic and help businesses and organizations better understand web traffic patterns.

show abstract

Section: Al Employed a Neuralprophet Model To Detectmentioning

confidence: 87%

Optimized Website Traffic Forecasting with Automatic Models and Optuna

Ahadi Ningrum,

Rifqi Mulyawan

2024

IIAI Letters on Informatics and Interdisciplinary Research

View full text Add to dashboard Cite

show abstract

“…In this context, Hernandez et al reviewed the SDG approaches proposed as an alternative to anonymization techniques for health domain applications [6]. Furthermore, there are also studies in which STSG has been researched and used [7][8][9][10][11][12][13][14].…”

Section: Related Workmentioning

confidence: 99%

“…This model is a probabilistic autoregressive (PAR) model, but its mathematical principles are still unpublished. In 2021, Hyun et al [13] proposed NeuralProphet, a neural network variation in the forecasting tool Prophet [17], as a method for STSG to create synthetic diabetic foot patients. In 2022, Li et al [14] presented the transformer-based time-series GAN (TTS-GAN) based on a transformer-encoder architecture.…”

Section: Related Workmentioning

confidence: 99%

Synthetic Subject Generation with Coupled Coherent Time Series Data

Larrea

Hernandez

Epelde

et al. 2022

The 8th International Conference on Time Series and Forecasting

View full text Add to dashboard Cite

show abstract

“…A sensor-based medical diabetic foot system proposed by [11] implements statistical methods, augmentation techniques, and a model to generate synthetic time series data. Results show that synthetic data follow the trends of real datasets.…”

Section: Introductionmentioning

confidence: 99%

Synthetic Health data Generation for Enhancement of Non-Invasive Diabetes AI-Based Prediction

Cruz Castañeda,

Bertemes Filho

2023

Preprint

View full text Add to dashboard Cite

Continuous glucose monitoring devices allow diabetes condition management. However, when limited data is available, one option is to increase their size by generating synthetic samples. From a homemade wearable prototype was created a real dataset with 18 instances and 53 attributes that capture characteristics of capillary and venous blood glucose, oxygen concentration, pulse rate, skin temperature, and 24 modules and 24 phases related to bio-impedance. The objective of this article is to generate synthetic datasets, and also it investigates the ideal features subset and optimal model for non-invasive diabetes prediction. Gaussian-Copulas (GC), conditional generative adversarial networks (CG), variational autoencoders, and Copula-GAN techniques' were used to generate five synthetic datasets. Experiments show that GC1 and GC2 datasets follow min/max boundaries and are not copies of the original data. Multilayer perceptron regressor outperformed (train and test) with 2.17, 2.51 in MAE; 9.29, 13.59 in MSE; 3.05, 3.69 in RMSE, and 0.95, 0.92 in R2 in GC1, and 2.64, 3.02 in MAE; 11.43, 15.11 in MSE; 3.38, 3.89 in RMSE, and 0.94, 0.92 in R2 in GC2 with eight features. Future work is necessary to explore autoencoder and generative architectures, datasets with diverse characteristics, and the effect of the number of features.

show abstract

Synthetic Data Generation System for AI-Based Diabetic Foot Diagnosis

Cited by 10 publications

References 21 publications

Optimized Website Traffic Forecasting with Automatic Models and Optuna

Optimized Website Traffic Forecasting with Automatic Models and Optuna

Synthetic Subject Generation with Coupled Coherent Time Series Data

Synthetic Health data Generation for Enhancement of Non-Invasive Diabetes AI-Based Prediction

Contact Info

Product

Resources

About