RESUMO
BACKGROUND: Acute kidney injury (AKI) after surgery appears to increase the risk of death in patients with liver cancer. In recent years, machine learning algorithms have been shown to offer higher discriminative efficiency than classical statistical analysis. AIM: To develop prediction models for AKI after liver cancer resection using machine learning techniques. METHODS: We screened a total of 2450 patients who had undergone primary hepatocellular carcinoma resection at Changzheng Hospital, Shanghai City, China, from January 1, 2015 to August 31, 2020. The AKI definition used was consistent with the Kidney Disease: Improving Global Outcomes. We included in our analysis preoperative data such as demographic characteristics, laboratory findings, comorbidities, and medication, as well as perioperative data such as duration of surgery. Computerized algorithms used for model development included logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGboost), and decision tree (DT). Feature importance was also ranked according to its contribution to model development. RESULTS: AKI events occurred in 296 patients (12.1%) within 7 d after surgery. Among the original models based on machine learning techniques, the RF algorithm had optimal discrimination with an area under the curve value of 0.92, compared to 0.87 for XGBoost, 0.90 for DT, 0.90 for SVM, and 0.85 for LR. The RF algorithm also had the highest concordance-index (0.86) and the lowest Brier score (0.076). The variable that contributed the most in the RF algorithm was age, followed by cholesterol, and surgery time. CONCLUSION: Machine learning algorithms are highly effective in discriminating patients at high risk of developing AKI. The successful application of machine learning models may help guide clinical decisions and help improve the long-term prognosis of patients.
RESUMO
BACKGROUND: Hepatic epithelioid angiomyolipoma (HEAML) is a rare liver disease and is easily misdiagnosed. Enhanced recognition of HEAML is beneficial to the differential diagnosis of rare liver diseases. CASE SUMMARY: We presented two cases of HEAML in Changzheng Hospital, Naval Medical University, and then collected and analyzed all reports about HEAML recorded in PubMed, MEDLINE, China Science Periodical Database, and VIP database from January 2000 to March 2018. A total of 409 cases of HEAML in 97 reports were collected, with a ratio of men to women of 1:4.84 and an age range from 12 years to 80 years (median 44 years). Among the patients with clinical symptoms mentioned, 61.93% (205/331) were asymptomatic, 34.74% (115/331) showed upper or right upper quadrant abdomen discomfort, while a few of them showed abdominal mass, gastrointestinal symptoms, low fever, or weight loss. The misdiagnosis rate of HEAML was as high as 40.34% (165/409) due to its nonspecific imaging findings. Most of the tumors were solitary and round in morphology, with clear boundaries. Ultrasound scan indicated low echo with internal nonuniformity and rich blood supply in most cases. Computer tomography/magnetic resonance imaging enhanced scan showed varied characteristics. The ratio of fast wash-in and fast wash-out, fast wash-in and slow wash-out, and delayed enhancement was roughly 4:5:1. A definite diagnosis of HEAML depended on the pathological findings of the epithelioid cells in lesions and the expression of human melanoma black 45, smooth muscle actin, melanoma antigen, and actin by immunohistochemical staining. HEAML had a relatively low malignant rate of 3.91%. However, surgical resection was the main treatment for HEAML, due to the difficulty diagnosing before operation. CONCLUSION: HEAML is a rare and easily misdiagnosed disease, and it should be diagnosed carefully, taking into account clinical course, imaging, pathological ,and immunohistochemical findings.