Table 3.

Overview of recent deep-learning development for classification.

Study	Data Set	Method/Model	Proposed Solution	Languages/Libraries/ Frameworks/ Tools/Software’s Used for Implementation and Simulation	Evaluation
Jakob Heime et al. [89]	Custom-developed (150 Samples)	Deep Neural Network	Deep Neural Network-based Customized Model	VIDI, COGNEX, Natick, MA, USA	Accuracy = 0.965% Threshold = 0.79% Sensitivity = 91.4% Specificity = 87.5%
Simukayi Mutasa et al. [91]	Digital Hand-Atlas (Samples)	Convolutional Neural Networks (CNNs)	Customized Convolutional Neural Networks with Inception Block	Python Tensor Flow v1.1 library, Ubuntu 16.04 workstation, NVIDIA TITAN X Pascal GPU.	Mean Absolute Error (MAE) = 0.536
Yagang WANG et al. [85]		GoogLeNet		Matlab	Accuracy = 94.4%
Toan Duc Bui et al. [95]	Public Dataset Digital Hand-Atlas (DHA) (1375 Samples)	Deep Convolutional Networks (DNNs)	Deep Convolutional Networks (DNNs) and Tanner Whitehouse (TW3) based Model	Not Mentioned	MAE = 0.59 RMS = 0.76
Jianlong Zhou et al. [97]	Digital Hand-Atlas	Convolutional Neural Networks	Convolutional Neural Networks and Transfer Learning Based Model	Not Mentioned	MAE = 0.72