Prediction and Clinically Important Factors of Acute Kidney Injury Non-recovery

Machine learning approaches toward an understanding of acute kidney injury: current trends and future directions

Acute kidney injury (AKI) is a significant health challenge associated with adverse patient outcomes and substantial economic burdens. Many authors have sought to prevent and predict AKI. Here, we comprehensively review recent advances in the use of artificial intelligence (AI) to predict AKI, and the associated challenges. Although AI may detect AKI early and predict prognosis, integration of AI-based systems into clinical practice remains challenging. It is difficult to identify AKI patients using retrospective data; information preprocessing and the limitations of existing models pose problems. It is essential to embrace standardized labeling criteria and to form international multi-institutional collaborations that foster high-quality data collection. Additionally, existing constraints on the deployment of evolving AI technologies in real-world healthcare settings and enhancement of the reliabilities of AI outputs are crucial. Such efforts will improve the clinical applicability, performance, and reliability of AKI Clinical Support Systems, ultimately enhancing patient prognoses.

A machine learning-based approach for predicting renal function recovery in general ward patients with acute kidney injury

BACKGROUND

Acute kidney injury (AKI) is a significant challenge in healthcare. While there are considerable researches dedicated to AKI patients, a crucial factor in their renal function recovery, is often overlooked. Thus, our study aims to address this issue through the development of a machine learning model to predict restoration of kidney function in patients with AKI.

METHODS

Our study encompassed data from 350,345 cases, derived from three hospitals. AKI was classified in accordance with the Kidney Disease: Improving Global Outcomes. Criteria for recovery were established as either a 33% decrease in serum creatinine levels at AKI onset, which was initially employed for the diagnosis of AKI. We employed various machine learning models, selecting 43 pertinent features for analysis.

RESULTS

Our analysis contained 7,041 and 2,929 patients' data from internal cohort and external cohort respectively. The Categorical Boosting Model demonstrated significant predictive accuracy, as evidenced by an internal area under the receiver operating characteristic (AUROC) of 0.7860, and an external AUROC score of 0.7316, thereby confirming its robustness in predictive performance. SHapley Additive exPlanations (SHAP) values were employed to explain key factors impacting recovery of renal function in AKI patients.

CONCLUSION

This study presented a machine learning approach for predicting renal function recovery in patients with AKI. The model performance was assessed across distinct hospital settings, which revealed its efficacy. Although the model exhibited favorable outcomes, the necessity for further enhancements and the incorporation of more diverse datasets is imperative for its application in real- world.

Using tumor habitat-derived radiomic analysis during pretreatment 18F-FDG PET for predicting KRAS/NRAS/BRAF mutations in colorectal cancer

BACKGROUND

To investigate the association between Kirsten rat sarcoma viral oncogene homolog (KRAS) / neuroblastoma rat sarcoma viral oncogene homolog (NRAS) /v-raf murine sarcoma viral oncogene homolog B (BRAF) mutations and the tumor habitat-derived radiomic features obtained during pretreatment 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) in patients with colorectal cancer (CRC).

METHODS

We retrospectively enrolled 62 patients with CRC who had undergone 18F-FDG PET/computed tomography from January 2017 to July 2022 before the initiation of therapy. The patients were randomly split into training and validation cohorts with a ratio of 6:4. The whole tumor region radiomic features, habitat-derived radiomic features, and metabolic parameters were extracted from 18F-FDG PET images. After reducing the feature dimension and selecting meaningful features, we constructed a hierarchical model of KRAS/NRAS/BRAF mutations by using the support vector machine. The convergence of the model was evaluated by using learning curve, and its performance was assessed based on the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis. The SHapley Additive exPlanation was used to interpret the contributions of various features to predictions of the model.

RESULTS

The model constructed by using habitat-derived radiomic features had adequate predictive power with respect to KRAS/NRAS/BRAF mutations, with an AUC of 0.759 (95% CI: 0.585-0.909) on the training cohort and that of 0.701 (95% CI: 0.468-0.916) on the validation cohort. The model exhibited good convergence, suitable calibration, and clinical application value. The results of the SHapley Additive explanation showed that the peritumoral habitat and a high_metabolism habitat had the greatest impact on predictions of the model. No meaningful whole tumor region radiomic features or metabolic parameters were retained during feature selection.

CONCLUSION

The habitat-derived radiomic features were found to be helpful in stratifying the status of KRAS/NRAS/BRAF in CRC patients. The approach proposed here has significant implications for adjuvant treatment decisions in patients with CRC, and needs to be further validated on a larger prospective cohort.

Kidney Function Trajectory within Six Months after Acute Kidney Injury Inpatient Care and Subsequent Adverse Kidney Outcomes: A Retrospective Cohort Study

Timing and extent of kidney function recovery after an acute kidney injury (AKI) episode are associated with chronic kidney disease onset and progression. This study aimed to categorize AKI recovery patterns within 6 months after index hospital discharge and associate them with kidney outcomes. This was a retrospective cohort study of 234,867 patients, hospitalized between 2010 and 2017, and classified as AKI or no AKI. Kidney function recovery from pre-hospitalization baseline within 1.5× serum creatinine (SCr) were evaluated at 3 and 6 months after hospital discharge and categorized as persistent non-recovery (PNR: SCr not recovered at 3 and 6 months), non-recovery (NR: SCr not recovered at 6 months), and recovery (SCr recovered at 6 months). A composite of incident chronic kidney disease, kidney replacement therapy, and estimated glomerular filtration rate reduction >30% from baseline and <15 mL/min/1.73 m2 was evaluated. Of 14,673 AKI surviving patients, 10.18% had PNR and 14.33% showed NR. Compared with no AKI, PNR and NR of AKI were associated with an increased risk of composite adverse outcomes (adjusted subdistribution hazard ratio (SHR) 4.55; 95% CI, 4.05−5.11; SHR, 3.54; 95% CI, 3.18−3.94, respectively). Patients with NR showed a greater risk of adverse outcomes than those with non-rapid recovery at 3 months after hospital discharge. The AKI recovery pattern within 6 months following inpatient care revealed an increasing continuum of risk of long-term adverse kidney outcomes. Risk stratification and a kidney function monitoring plan at discharge are needed to improve post-AKI care.