|
Author | Type of feature extracted | ML segmentation technique | No. of data and details | Performance (accuracy %) |
|
Begum and Razak [52] | Morphological operations erosion, dilation, opening, and closing of nuclei | SVM | Not mentioned | 83% |
|
Jothi et al. [53] | Morphological, wavelet, color, texture, and statistical features, and other features | Naıve Bayes, linear discriminant analysis, K-nearest neighbor, support vector machine, decision tree, and ensemble random under sampling boost | 300 | (60%–100%) |
|
Gajul and Shelke [54] | Not mentioned | K-mean clustering and morphological operations | 40 | — |
|
Vogado et al. [22] | Not identified | Automatic feature extraction | Three datasets | 99% |
|
Agaian et al. [55] | Color, texture, shape, and Hausdorff dimension feature | Using k-means clustering | 80 | 98% |
|
Negm et al. [51] | Geometric, color, texture, and size feature of blast | Using k-means clustering | 75 | 99.5% |
|
Su et al. [39] | Color and morphology features | k-means cluster and constructing a cell image by hidden Markov random field | 61 | 96% |
|
Goutam and Sailaja [56] | LDP feature | Using k-means clustering | 90 | 98% |
|
Shankar et al. [57] | Color, shape, and texture | Threshold by using the Zack algorithm | 33 | 96% |
|
Viswanathan [46] | Morphological contour (edge detection, erosion, and dilation) | Fuzzy c-means | — | 98% |
|
Patel and Mishra [58] | Geometric, color, texture, and size feature of cell | K-mean clustering and the Zack algorithm | 7 | 93% |
|
Zhao et al. [12] | Morphological operation and granularity feature are selected automatically | CNN and SVM | 9 | 94% |
|
Karthikeyan and Poornima [59] | Geometrical, texture, and color | Fuzzy c-means | 19 | 90% |
|
MoradiAmin et al. [37] | Geometrical and statistical feature | Fuzzy c-means | 21 | 98% |
|
Rawat et al. [60] | Morphological operation | Global thresholding and morphological opining | 260 | (79%–95.4%) |
|
Mishra et al. [61] | Texture and color | Marker-controlled watershed segmentation | 190 | 96% |
|
Bhattacharjee and Saini [36] | Morphological operation | Morphological operations erosion, dilation, opening, and closing | 120 | 96% |
|
Khobragade et al. [62] | Geometrical and statistical | Otsus’s thresholding and Sobel operator | Not mentioned | 90% |
|
Patil and Raskar [41] | Color, shape, and texture | Thresholding by using Otsu’s method | Not mentioned | Not mentioned |
|
Rawat et al. [34] | Shape features | Global thresholding and morphological opining | 420 | 96.75% |
|
Neelam et al. | Texture features | K-mean clustering followed by expectation maximization algorithm | Not mentioned | 80% |
|
Singh et al. [63] | Shape and texture features | ANN | ALL-IDB (no: 108) | 97.2% |
|
Singhal and Singh [64] | Texture features | SVM | ALL-IDB (no: 260) | 93.8% |
|
Zhang et al. [65] | Shape features | Fuzzy system | Local (not mentioned) | Not mentioned |
|
Neoh et al. [66] | Shape, texture, and color features | Dempster–Shafer | ALL-IDB (no: 180) | 96.7% |
|
Amin et al. [67] | Shape and texture features | SVM | Local (no: 21) | 97% |
|
Viswanathan [46] | Shape, color, and texture features | Fuzzy c-means classifier | ALL-IDB (no: 108) | 98.0% |
|
Bhattacharjee and Saini [36] | Shape features | ANN | ALL-IDB (no: 120) | 95.2% |
|
ElDahshan et al. [68] | Not mentioned | Field | ALL-IDB (no: 300) | Not mentioned |
|
Rawat et al. [60] | Shape and texture features | SVM | ALL-IDB (no: 196) | 89.8% |
|
Putzu et al. [35] | Shape, color, and texture features | SVM | ALL-IDB (no: 267) | 92.0% |
|
Mohapatra et al. [70] | Shape and texture features | Ensemble classifier | Local dataset (no: 104) | 94.7% |
|
Nasir et al. [71] | Shape and color features | MLP_BR | Local dataset (no: 230) | 95.7% |
|
Mohapatra et al. [72] | Shape and texture features | ANN | Local dataset (no: 100) | Not mentioned |
|
Madhloom et al. [73] | Shape and texture features | kNN clustering | Local dataset (no: 260) | 92.5% |
|
Pedreira et al. [74] | Multiple clinical and laboratorial features | ANN | Local dataset (no: 189) | 98.2% |
|