Table 2. Comparison of cell death identification studies.

Table reporting all studies on cell death classification based on machine learning. For each study, we included the reported classification accuracy, the experimental conditions of the studies, the target input of the classifier, and the capability of performing detection on static frames or microscopy timelapses. Met conditions are indicated with a green check. Moreover, for each study we reported the architecture of the classifier and the number of apoptotic cells in the training set. NA stands for not available and indicates that the information is not reported in the study.

Study	Input of the classifier	Reported classification accuracy	In vitro	In vivo	DetectionIn frame	Detection in movies	Classifier architecture	N cell death
Our	Frame sequence	98.27%	✓	✓	✓	✓	Conv-Transformer	13,120
Jin et al., 2022	Frame	93%	✓	✘	✘	✘	Logistic regression	NA
Verduijn et al., 2021	Frame	87%	✓	✘	✘	✘	VGG-19	19,339
Kabir et al., 2022	Frame sequence	93%	✓	✘	✘	✘	ResNet101-LSTM	3172
La Greca et al., 2021	Frame	96.58%	✓	✘	✘	✘	ResNet50	11,036
Mobiny et al., 2020	Frame sequence	93.8%	✓	✘	✘	✘	CapsNet-LSTM	41,000
Kranich et al., 2020	Frame	93.2%	✓	✘	✘	✘	CAE-RandomForest	27,224
Vicar et al., 2020	Frame sequence	NA	✓	✘	✓	✓	biLSTM	1745
Jimenez-Carretero et al., 2018	Frame	NA	✓	✘	✓	✘	R-CNN	255,215