 
                This platform aims to establish an AD prediction model for MCI subjects using routine laboratory test data and machine-learning methods. With a large population of 43,579 patients with AD and 4,135 patients with MCI, and 210 biomarkers gathered from routine laboratory tests accumulated from 2000 to 2019 in Hong Kong, we were able to generate classification models for AD prediction by ensemble various machine learning algorithms models, such as linear model, tree model and support vector machine (SVM). Additionally, analyses were further grouped by gender (male and female) and age (65-74, 75-89) to improve the accuracy of early screening for AD. The probability represents the model’s prediction of MCI conversion to AD. The threshold is 0.5. If the probability of the model is 0.5 or less, it is MCI. Otherwise, it is AD.
The platform uses a minimum of 5 biomarkers and a maximum of 31 biomarkers for different age and gender. 31 biomarkers including:
1 Biliary function items: Bilirubin (total).
2 Renal function items: Creatinine, Urea.
2 Liver function items: Alanine aminotransferase, Alkaline phosphatase.
4 Microelements items: Potassium, Sodium, Calcium, Phosphate.
6 Protein items: Albumin, Globulin, Protein (total), Haemoglobin (blood), MCH, MCHC.
16 Blood Cell items: Eosinophil (%), Basophil (%), Neutrophil (%), Lymphocyte (%), Monocyte (%), Eosinophil (absolute), Lymphocyte (absolute), Basophil (absolute), Neutrophil (absolute), Monocyte (absolute), WBC, RBC, RDW(%), MCV, Platelet, HCT. TIPS: Red blood cell (RBC), mean corpuscular volume (MCV), Mean corpuscular hemoglobin (MCH), Mean Corpuscular Hemoglobin Concentration (MCHC), White blood cell (WBC), Red blood cell distribution width (RDW), Hematocrit (HCT). Eosinophil (%), Basophil (%), Neutrophil (%), Lymphocyte (%), Monocyte (%), Eosinophil (absolute), Lymphocyte (absolute), Basophil (absolute), Neutrophil (absolute), Monocyte (absolute), WBC, RBC, RDW(%), MCV, Platelet, HCT.
@article{cao2024visible,
  title={Visible and Clear: Finding Tiny Objects in Difference Map},
  author={Cao, Bing and Yao, Haiyu and Zhu, Pengfei and Hu, Qinghua},
  journal={arXiv preprint arXiv:2405.11276},
  year={2024}
}
		<pre>
@ARTICLE{Gao22LUSS,
author={Gao, Shanghua and Li, Zhong-Yu and Yang, Ming-Hsuan and Cheng, Ming-Ming and Han, Junwei and Torr, Philip},
title={Large-scale Unsupervised Semantic Segmentation},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
year={2023},
volume={45},
number={6},
pages={7457-7476},
doi={10.1109/TPAMI.2022.3218275}
}
</pre>