Table 2.
Predictive models employed in staging, treatment and prognosis of endometrial cancer
| Author | Country | Study design | Objective | Outcome |
|---|---|---|---|---|
| Lowe et al.50 | United States | Non-randomized control clinical trial | To report perioperative outcomes and learning curve characteristics from a multi-institutional experience with robotic-assisted surgical staging for endometrial cancer. | Robotic technology may be an excellent intervention for EC staging, but it further requires its comparison with laparoscopy and laparotomy |
| Cardenas-Goicoechea et al.51 | United States | Retrospective review | To compare the survival outcomes of women with endometrial cancer managed by robotic and traditional laparoscopic-assisted surgery | There were no significant differences in survival (3-year survival 93.3% and 93.6%), DFS (3-year DFS 83.3% and 88.4%), and tumour recurrence (14.8% and 12.1%) for robotic and laparoscopic groups, respectively |
| Göçmen et al.52 | Turkey | Prospective study | To compare the results of patients on whom staging was applied by robotic-assisted laparoscopic surgery and laparotomy for endometrial cancer | The outcome of this study was in favour of robotic-assisted surgery with its advantages of less hospital stay duration, low blood loss, and less frequency of lymph node dissection |
| Lecointre et al.54 | France | Systematic review | To investigate the contribution of radiomics on the radiological preoperative assessment of patients with EC; and to establish a simple and reproducible AI Quality Score applicable to machine learning and deep learning studies | Preliminary data indicates that these new technologies, when combined with human intelligence, can address some of the clinical problems, even though there is not enough proof to support the use of radiomics in the treatment of endometrial cancer. |
| Fell et al.55 | Scotland | Cross sectional | To categorize endometrial biopsy whole-slide images (WSI) from digital pathology as either “malignant”, “other or benign” or “insufficient” with the help of AI | The final model accurately classifies 90% of all slides correctly and 97% of falls in the malignant class, suggesting that the use of AI in screening whole cell slides for determining their cytology is enough to implement this for the reduction of pathologists’ workload |
| Markis et al.56 | Greece | Cross sectional | To investigate the efficacy of an artificial neural network based on multi-layer perceptron (ANN–MPL) to discriminate between benign and malignant endometrial nuclei and lesions in cytological specimens. | For the case classification based on the numeric classifier, the overall accuracy was 90.87%, the specificity 93.03%, and the sensitivity 87.79%; the indices for the percentage classifier were 95.91%, 93.44%, and 99.42%, respectively. These invented computerized systems based on ANNs can be helpful for the cytological classification of endometrial nuclei and lesions with adequate sensitivity and specificity |
| Bell et al.57 | United States | Retrospective chart review | To compare hysterectomy and lymphadenectomy completed via robotic assistance, laparotomy, and laparoscopy for endometrial cancer staging with respect to operative and perioperative outcomes, complications, adequacy of staging, and cost | Robotic hysterectomy provides better node retrieval to laparotomy and laparoscopic procedures if performed by a skilled laparoscopic surgeon. Robotics provides the patient with a speedier recovery and reduces the risk of postoperative morbidity. Moreover, the average cost for hysterectomy and staging was highest for laparotomy, followed by robotic, and least for standard laparoscopy |
| Feng et al.58 | China | Retrospective study | To develop a deep learning (DL) model for prediction of lymph node metastasis (LNM) based on hematoxylin and eosin (HE)-stained histopathological images of EC. The model was validated using external data | It was discovered that a novel DL-based biomarker, especially for patients in the early stages of staging, might predict metastatic status with enhanced accuracy after being trained on various histological subtypes of EC slides |
| Mysona et al.59 | United States | Retrospective study | To determine the utility of a clinical calculator to predict the benefit of chemotherapy in stage IA uterine papillary serous cancer (UPSC) | it was discovered that a low-risk group would not benefit from chemotherapy when the relative benefits of the treatment were evaluated |
AI, artificial intelligence; DFS, disease-free survival; EC, endometrial carcinoma.