The early identification of toxic paraquat (PQ) poisoning in patients is critical to ensure timely and accurate prognosis. Although plasma PQ concentration has been reported as a clinical indicator of PQ poisoning, it is not commonly applied in practice due to the inconvenient necessary instruments and operation. In this study, we explored the use of blood routine indexes to identify the degree of PQ toxicity and/or diagnose PQ poisoning in patients via machine learning approach. Specifically, we developed a method based on support vector machine combined with the feature selection technique to accurately predict PQ poisoning risk status, then tested the method on 79 (42 male and 37 female; 41 living and 38 deceased) patients. The detection method was rigorously evaluated against a real-world data set to determine its accuracy, sensitivity and specificity. Feature selection was also applied to identify the factors correlated with risk status, and the results showed that there are significant differences in blood routine indexes between dead and living PQ-poisoned individuals (p-value < 0.01). Feature selection also showed that the most important correlated indexes are white blood cell and neutrophils. In conclusion, the toxicity or prognosis of PQ poisoning can be preliminarily ascertained by blood routine testing without PQ concentration data, representing an additional tool and innovative approach to assess the prognosis of PQ poisoning.of the most commonly used herbicides in the world, is considerably more toxic to human beings than other herbicidal agents. Acute ingestion of 7-8 mL PQ can cause severe liver, lung and kidney issues and even heart failure leading to death [1]; the mortality of PQ poisoning is 60-70% [2]. Many people have been poisoned by PQ throughout many years and across the world, although the problem is especially serious in developing countries. Early prognosis of PQ poisoning is significant for both patients and doctors [3]. Studies have shown that the mortality rate of PQ poisoning is highly correlated with ingested PQ amount or PQ concentrations in plasma and urine (within 4-24 hr elapsed from ingestion) [4,5]. Hsu et al. [6] found that early haemoperfusion (HP) after PQ exposure may reduce the mortality specifically and that HP <4.0 or <5.0 hr can reduce the relative risk of mortality by approximately 81% or 51%, respectively.Although PQ concentration has proven effective in terms of PQ poisoning diagnosis and prognosis, it is seldom applied in clinics [7,8] because it must be determined by HPLC or LC-MS, which requires highly accurate equipment [9,10].Although sodium dithionite, sodium hydroxide and spectrophotometry can be used to determine PQ concentration, they are simple qualitative assays and cannot provide the accurate level of PQ concentration. Consequently, PQ concentration is generally not available as a routine test option in hospitals.The blood routine test (BRT), also named complete blood count, which includes many indexes such as white blood cell (WBC), red blo...