Table 3.

Published studies using artificial intelligence for predicting the main kidney transplant-related complications

Predicted outcome	References	Data size	Predictive factors	Used algorithms	Performance
Delayed graft function	Jen et al. 2021 [57]	Not mentioned	Donor: age, height, weight, ethnicity, serum creatinine, BUN, hypertension, cardiac death, cause of death, and CIT	ANN	AUC: 0.7595
	Konieczny et al. 2021 [58]	157 KTs	Donor: eGFR, BMI, age, gender, and kidney donor profile index Recipient: BMI Recipient–donor weight difference	RF ANN	Accuracy: 84.38–93.75% AUC: 0.84–0.91
	Bae et al. 2020 [59]	133,431 KTs	Recipient: age, gender, race, time on dialysis, cause of ESRD, panel reactive antibody, BMI, previous transplant, CIT Donor: age, gender, race, serum creatinine, cause of death	GB RF	C-statistic: C = 0.591 [0.581–0.601] C = 0.579 [0.569–0.589]
	Kawakita et al. 2020 [60]	61,220 KTs	Recipient: pretransplant dialysis, serum creatinine, BMI, days on waiting list, race, gender, diabetes and initial waitlist status Donor: age, type, serum creatinine, BUN, BMI, hypertension, cause of death, medication Transplant: biopsy, CIT, kidney pump	RF LR XGB ANN	AUC: 0.70–0.735
	Costa et al. (2020) [61]	443 KTs	Donor maintenance-related variables: arterial blood gas pH, serum sodium, blood glucose, urine output, mean arterial pressure, vasopressor use, and reversed cardiac arrest	DT, ANN SVM	AUC: 0.784—0.886
	Decruyenaere et al. 2015 [62]	55,044 KTs	Not mentioned	RF LR SVM XGB ANN	AUC: 0.705–0.735
	Li et al. 2010 [63]	1228 KTs	Recipient: age, gender, socio-economic level, primary diagnosis, functional status, medical condition, previous KT, hospitalization 90 days prior to KT, pre-transplant dialysis Donor: type, BMI, transplant side Procedure: surgeon, WIT, CIT, anastomotic time, incidental tumor Post-transplant data: urine output, complications	Bayesian network	AUC: 0.824–0.967
	Brier et al. 2003 [64]	304 KTs	Donor: body surface area CIT Recipient: Age, race, height and weight HLA matching (AB loci, DR loci)	ANN	AUC: 0.668
	Shoskes et al. 1999 [65]	100 KTs	Donor: age, cause of death, history of hypertension, inotrope use, urine output, and both initial and terminal serum creatinine	ANN	Not mentioned
Acute rejection	Shaikhina et al. 2019 [51]	80 KTs	Recipient: age, gender, HLA Class I and II, years on dialysis, total IgG subclass [1–4] levels, highest IgG DSA level, and number of previous transplants Cytometery cross-match, HLA mismatches, number of class II HLA-DR mismatches Donor type	DT RF	AUC (DT): 0.849 AUC (RF): 0.819
	Tapak et al. 2017 [66]	378 KTs	Donor: age, gender, type, familial relationship, blood group, and BMI Recipient: age, gender hemoglobin level, blood group, duration of dialysis before transplantation, CIT, creatinine level at discharge, left or right kidney, immunosuppressive drugs used, duration of hospitalization, urine output during the first 24 h, and occurrence of acute or hyperacute rejection	ANN LR	AUC (ANN): 0.88 AUC (LR): 0.75
	Esteban et al. 2016 [67]	Not mentioned	Prescribed medication (Cyclosporin, Furosemide…) Creatinine (high/normal/low) Leukocytes (high/normal/low)	Recurrent Neural Networks	AUC: 0.082
	Hummel et al. 2010 [68]	145 KTs	Consultation time after transplantation, tacrolimus dose, induction therapy, renal initial function, donor type, CMV in the recipient, diuresis, temperature increase, edema, tremor, urea dosage, serum creatinine, blood glucose, leukocyte count, lymphocyte count, platelet count, AT average, and histocompatibility	ANN	AUC: 0.7568
	Petrovsky et al. 2002 [69]	1542 KTs	Donor: age, gender, graft side Recipient: age, CMV and EBV status, other organ transplants, transfusions Institutions: referring, hospital, state) Transplant: total ischemia, kidney preservation HLA matching (HLA-A, -B, -DR, and DQ)	ANN	Accuracy: 71.7%
	Abdolmaleki et al. 1997 [70]	35 KTs	Urine output, blood urea nitogen, serum creatinine, perfusion index, renal peak to plateau ratio …	ANN	R:0.78
Long-term graft survival	Badrouchi et al. 2021 [19]	407 KTs	Recipient: history of hypertension, history of transfusion, duration on dialysis before KT, acute kidney injury post-KT, acute rejection, CMV infection, length of the 1st hospitalization, 3-month eGFR, MMF therapy Donor’s age	XGB	AUC: 0.897 Se: 91% Sp: 87%
	Loupy et al. 2019 [25]	4000 KTs	Time of post-transplant risk evaluation, eGFR, proteinuria, histological parameters (interstitial fibrosis,tubular atrophy, glomerulitis, peritubular capillaritis, interstitial inflammation, tubulitis and transplant glomerulopathy) and DSA	Cox regression with bootstrapping for validation	C index: 0.819 95% CI [0.799–0.839]
	Nematollahi et al. 2017 [71]	717 KTs	Recipient: age, cause of ESRD, duration on dialysis before KT, serum creatinine at discharge, and hypertension after KT Donor: age, blood group	SVM	Sn: 97.3% Sp: 26.1% Accuracy: 85.9% AUC: 0.769
	Shahmoradi et al. 2016 [72]	513 KTs	Donor age, donor gender, recipient age, recipient gender, cause of ESRD, dialysis, duration on dialysis, panel test, BMI, donor type	ANN DT Bayesian network	Sn: 78.1–90.8% Sp: 52.0–65% Accuracy: 83.3- 87.2%
	Lofaro et al. 2016 [73]	80 KTs	Recipient age, number of transplants, 6-month eGFR, 6-month 24-h urine protein excretion, 6-month serum hemoglobin and 6-month hematocrit	DT	Sn: 62.5% Sp: 92.8% AUC: 0. 847%
	Greco et al. (2010) [74]	194 KTs	Recipient BMI, DGF, acute rejection episode and chronic allograft nephropathy	DT	Sn: 88.2% Sp: 73.8%
	Akl et al. 2008 [75]	1900 KTs	Recipient age, transfusions, Haplotype, time to diuresis, total steroid dose (first 3 months), immunosuppression, acute tubular necrosis, and acute rejection episodes (first 3 months) Total HLA MM, HLA DR MM, Donor: age, sibling/related/unrelated donor	ANN	Sn: 88.4% Sp: 73.2% Accuracy: 95% AUC: 0.88
	Lin et al. 2008 [55]	57,383 KTs	Recipient: age, gender, race, height, weight, cause of ESRD, history of hypertension, diabetes or CV disease, duration between date of current KT and failure date of the previous KT (if applicable), dialysis modality, predominant dialysis modality, and primary source of payment for treatment Donor: type, age, gender, race, height, weight and cause of death Number of matched HLA antigens, CIT and procedure type	ANN	AUC: 0.73–0.82
	Krikov et al. 2007 [76]	92,844 KTs	Recipient: race, gender, age, height, weight, prior KT, number of KTs, time on waiting list, predominant RRT modality, PD before KT, number of RRT modalities used before transplant, specific combination of RRT modalities, recipient comorbidity score, history of CV disease, history of unstable angina, history of diabetes, history of hypertension, VHB, VHC, peak and most recent level of panel reactive antibodies and primary source of payment for medical services Donor: race, gender, age, height, weight, donor type (living or deceased)	DT	AUC: 0.62–0.983

KT kidney transplantation, BUN blood urea nitrogen, CIT cold ischemia time, eGFR estimated glomerular filtration rate, BMI body mass index, ESRD end-stage renal disease, WIT warm ischemia time, HLA human leukocyte antigen, MM mismatch, MMF mycophenolate mofetil, CV cardiovascular disease, PD peritoneal dialysis, RRT renal replacement therapy, VHC viral hepatitis C, VHB viral hepatitis B, AUC area under the curve, Se sensitivity, Sp specificity