Table 1.

Summary of literature survey on sickle cell disease.

References	Dataset	Best Classifiers	Highest Accuracy	Key Contributions	Limitations
[43]	erythrocytesIDB	DT RF	94.44 % 95.06 %	Chose the best ML classification method to detect SCD.	Limited Dataset and ML-centric implementation.
[45]	SCD dataset	LEVNN	99.1 %	The preprocessing of the medical time-varying data stream was improved.	Addressed a few architectures which limited their scale and scope.
[46]	Collected	Model 2 + SVM in Scenario 4	99.98 %	Classified 3 types of RBC using 3 deep-learning models.	Data imbalance. Data Quality Trade-off. Does not provide details about the sources or potential issues related to data quality, reliability and variability about Dataset 3 which is from different websites and internet searches.
[50]	Collected	NB KNN	98.212 % 98.87 %	Automatic classification of cell motions in videos.	Uniform downsampling was poor and unreliable.
[52]	Collected	CNN C-RNN	93.5 % 97 %	Hybrid segmentation made detection automatic.	Excluded sickle cell images overlapping with RBC.