A machine learning approach for distinguishing uterine sarcoma from leiomyomas based on perfusion weighted MRI parameters

Validation of biomarkers and clinical scores for the detection of uterine leiomyosarcoma: a case-control study with an update of pLMS

BACKGROUND

The diagnosis of rare uterine leiomyosarcoma (uLMS) remains a challenge given the high incidence rates of benign uterine tumors such as leiomyoma (LM). In the last decade, several clinical scores and blood serum markers have been proposed. The aim of this study is to validate and update the pLMS clinical scoring system, evaluating the accuracy of the scoring system by Zhang et al. and examining the discriminatory ability of blood markers such as serum lactate dehydrogenase (LDH), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR).

METHODS

In a case-control study, 90 new uLMS from the DKSM consultation registry and 659 prospectively recruited LM cases from the Saarland University Hospital were used for validation. Welch's t-test and Hedges' g were used to evaluate blood markers and optimal thresholds and diagnostic odds ratios were calculated. Scoring systems were compared using receiver operating characteristics and proposed diagnostic cut-offs were reviewed. Missing values were imputed by random forest imputation to create the updated scoring system 'pLMS2' using penalized logistic regression based on the pooled data sets of 384 uLMS and 1485 LM.

RESULTS

pLMS achieved an AUC of 0.97 on the validation data, but sensitivity and specificity varied at the proposed thresholds due to a shift in the score distributions. 43 uLMS and 578 LM were included in the comparison of pLMS with Zhang's scoring system, with pLMS being superior (AUC 0.960 vs 0.845). LDH, NLR, and PLR achieved a diagnostic odds ratios of 18.03, 8.64 and 4.81, respectively. pLMS2 is based on subscores for menopausal status interacting with age, tumor diameter, intermenstrual bleeding, hypermenorrhea, dysmenorrhea, postmenstrual bleeding, rapid tumor growth, and suspicious sonography.

CONCLUSIONS

Validation of the pLMS shows stable discriminatory ability as expressed by AUC, although caution should be taken with cut-off values, as sensitivity and specificity may vary. Data collection of the updated clinical score pLMS2 remains simple and convenient, with no additional cost. The proposed thresholds of 1.5 and 5.5 can be used as a guide to avoid unnecessary or inappropriate surgery and to make the use of further diagnostic measures cost-effective. LDH, NLR and PLR provide further evidence to differentiate uLMS from LM in conjunction with clinical data.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in differentiation of soft tissue sarcoma from benign lesions: a systematic review of literature

OBJECTIVE

To systematically review the literature assessing the role of Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) in the differentiation of soft tissue sarcomas from benign lesions.

MATERIALS AND METHODS

A comprehensive literature search was performed with the following keywords: multiparametric magnetic resonance imaging, DCE-MR perfusion, soft tissue, sarcoma, and neoplasm. Original studies evaluating the role of DCE-MRI for differentiating benign soft-tissue lesions from soft-tissue sarcomas were included.

RESULTS

Eighteen studies with a total of 965 imaging examinations were identified. Ten of twelve studies evaluating qualitative parameters reported improvement in discriminative power. One of the evaluated qualitative parameters was time-intensity curves (TIC), and malignant curves (TIC III, IV) were found in 74% of sarcomas versus 26.5% benign lesions. Six of seven studies that used the semiquantitative approach found it relatively beneficial. Four studies assessed quantitative parameters including Ktrans (contrast transit from the vascular compartment to the interstitial compartment), Kep (contrast return to the vascular compartment), and Ve (the volume fraction of the extracellular extravascular space) in addition to other parameters. All found Ktrans, and 3 studies found Kep to be significantly different between sarcomas and benign lesions. The values for Ve were variable. Additionally, eight studies assessed diffusion-weighted imaging (DWI), and 6 of them found it useful.

CONCLUSION

Of different DCE-MRI approaches, qualitative parameters showed the best evidence in increasing the diagnostic performance of MRI. Semiquantitative and quantitative approaches seemed to improve the discriminative power of MRI, but which parameters and to what extent is still unclear and needs further investigation.

The automatic diagnosis artificial intelligence system for preoperative magnetic resonance imaging of uterine sarcoma

OBJECTIVE

Magnetic resonance imaging (MRI) is efficient for the diagnosis of preoperative uterine sarcoma; however, misdiagnoses may occur. In this study, we developed a new artificial intelligence (AI) system to overcome the limitations of requiring specialists to manually process datasets and a large amount of computer resources.

METHODS

The AI system comprises a tumor image filter, which extracts MRI slices containing tumors, and sarcoma evaluator, which diagnoses uterine sarcomas. We used 15 types of MRI patient sequences to train deep neural network (DNN) models used by tumor filter and sarcoma evaluator with 8 cross-validation sets. We implemented tumor filter and sarcoma evaluator using ensemble prediction technique with 9 DNN models. Ten tumor filters and sarcoma evaluator sets were developed to evaluate fluctuation accuracy. Finally, AutoDiag-AI was used to evaluate the new validation dataset, including 8 cases of sarcomas and 24 leiomyomas.

RESULTS

Tumor image filter and sarcoma evaluator accuracies were 92.68% and 90.50%, respectively. AutoDiag-AI with the original dataset accuracy was 89.32%, with 90.47% sensitivity and 88.95% specificity, whereas AutoDiag-AI with the new validation dataset accuracy was 92.44%, with 92.25% sensitivity and 92.50% specificity.

CONCLUSION

Our newly established AI system automatically extracts tumor sites from MRI images and diagnoses them as uterine sarcomas without human intervention. Its accuracy is comparable to that of a radiologist. With further validation, the system could be applied for diagnosis of other diseases. Further improvement of the system's accuracy may enable its clinical application in the future.

A human-interpretable machine learning pipeline based on ultrasound to support leiomyosarcoma diagnosis

The preoperative evaluation of myometrial tumors is essential to avoid delayed treatment and to establish the appropriate surgical approach. Specifically, the differential diagnosis of leiomyosarcoma (LMS) is particularly challenging due to the overlapping of clinical, laboratory and ultrasound features between fibroids and LMS. In this work, we present a human-interpretable machine learning (ML) pipeline to support the preoperative differential diagnosis of LMS from leiomyomas, based on both clinical data and gynecological ultrasound assessment of 68 patients (8 with LMS diagnosis). The pipeline provides the following novel contributions: (i) end-users have been involved both in the definition of the ML tasks and in the evaluation of the overall approach; (ii) clinical specialists get a full understanding of both the decision-making mechanisms of the ML algorithms and the impact of the features on each automatic decision. Moreover, the proposed pipeline addresses some of the problems concerning both the imbalance of the two classes by analyzing and selecting the best combination of the synthetic oversampling strategy of the minority class and the classification algorithm among different choices, and the explainability of the features at global and local levels. The results show very high performance of the best strategy (AUC = 0.99, F1 = 0.87) and the strong and stable impact of two ultrasound-based features (i.e., tumor borders and consistency of the lesions). Furthermore, the SHAP algorithm was exploited to quantify the impact of the features at the local level and a specific module was developed to provide a template-based natural language (NL) translation of the explanations for enhancing their interpretability and fostering the use of ML in the clinical setting.

Artificial intelligence-based risk stratification, accurate diagnosis and treatment prediction in gynecologic oncology

As data-driven science, artificial intelligence (AI) has paved a promising path toward an evolving health system teeming with thrilling opportunities for precision oncology. Notwithstanding the tremendous success of oncological AI in such fields as lung carcinoma, breast tumor and brain malignancy, less attention has been devoted to investigating the influence of AI on gynecologic oncology. Hereby, this review sheds light on the ever-increasing contribution of state-of-the-art AI techniques to the refined risk stratification and whole-course management of patients with gynecologic tumors, in particular, cervical, ovarian and endometrial cancer, centering on information and features extracted from clinical data (electronic health records), cancer imaging including radiological imaging, colposcopic images, cytological and histopathological digital images, and molecular profiling (genomics, transcriptomics, metabolomics and so forth). However, there are still noteworthy challenges beyond performance validation. Thus, this work further describes the limitations and challenges faced in the real-word implementation of AI models, as well as potential solutions to address these issues.

A Comprehensive Review of the Role of Artificial Intelligence in Obstetrics and Gynecology

The exponential growth of artificial intelligence (AI) has fascinated its application in various fields and so in the field of healthcare. Technological advancements in theories and learning algorithms and the availability of processing through huge datasets have created a breakthrough in the medical field with computing systems. AI can potentially drive clinicians and practitioners with appropriate decisions in managing cases and reaching a diagnosis, so its application is extensively spread in the medical field. Thus, computerized algorithms have made predictions so simple and accurate. This is because AI can proffer information accurately even to many patients. Furthermore, the subsets of AI, namely, machine learning (ML) and deep learning (DL) methods, have aided in detecting complex patterns from huge datasets and using such patterns in making predictions. Despite numerous challenges, AI implementation in obstetrics and gynecology is found to have a spellbound development. Therefore, this review propounds exploring the implementation of AI in obstetrics and gynecology to improve the outcomes and clinical experience. In that context, the evolution and progress of AI, the role of AI in ultrasound diagnosis in distinct phases of pregnancy, clinical benefits, preterm birth postpartum period, and applications of AI in gynecology are elucidated in this review with future recommendations.

The role of multiparametric MRI in differentiating uterine leiomyosarcoma from benign degenerative leiomyoma and leiomyoma variants: a retrospective analysis

AIM

To assess qualitative and quantitative magnetic resonance imaging (MRI) factors that can help distinguish leiomyosarcoma (LMS) from benign degenerative leiomyoma (BDL) and leiomyoma variants (LV) and assess the interobserver agreement for the proposed quantitative factors.

MATERIALS AND METHODS

Retrospective analysis of all histopathology proven cases of LV, BDL, and LMS with a preoperative MRI was performed. Twenty-seven cases were included (five LMS, three LV, and 19 BDL) with each case independently read by a pair of radiologists. Lesion size, margins, presence or absence of degeneration, necrosis, and haemorrhage were assessed on MRI along with quantitative factors such as mean T2-weighted (W) and T1W signal intensity, T1W signal heterogeneity, diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) ratios as well as dynamic contrast enhancement (DCE) characteristics along with the presence or absence of lymphadenopathy and extra-uterine and peritoneal spread. Mean and standard deviation for quantitative variables and frequency with percentages for qualitative variables were assessed.

RESULTS

Infiltrative margins were seen exclusively in the LMS group (n=1), with the remaining LMS cases showing lobulate or rounded smooth margins similar to BDL or LV. A high T2W signal <25% was seen exclusively in the BDL group (n=8). The presence of concomitant necrosis and haemorrhage was seen exclusively in the LMS group (n=2). Quantitative MRI had good inter-reader correlation but was not significantly different between the LMS, BDL, and LV groups.

CONCLUSION

LMS, BDL, and LV may have overlapping features on multiparametric MRI making differentiation difficult.

Construction of machine learning-based models for cancer outcomes in low and lower-middle income countries: A scoping review

Background

The impact and utility of machine learning (ML)-based prediction tools for cancer outcomes including assistive diagnosis, risk stratification, and adjunctive decision-making have been largely described and realized in the high income and upper-middle-income countries. However, statistical projections have estimated higher cancer incidence and mortality risks in low and lower-middle-income countries (LLMICs). Therefore, this review aimed to evaluate the utilization, model construction methods, and degree of implementation of ML-based models for cancer outcomes in LLMICs.

Methods

PubMed/Medline, Scopus, and Web of Science databases were searched and articles describing the use of ML-based models for cancer among local populations in LLMICs between 2002 and 2022 were included. A total of 140 articles from 22,516 citations that met the eligibility criteria were included in this study.

Results

ML-based models from LLMICs were often based on traditional ML algorithms than deep or deep hybrid learning. We found that the construction of ML-based models was skewed to particular LLMICs such as India, Iran, Pakistan, and Egypt with a paucity of applications in sub-Saharan Africa. Moreover, models for breast, head and neck, and brain cancer outcomes were frequently explored. Many models were deemed suboptimal according to the Prediction model Risk of Bias Assessment tool (PROBAST) due to sample size constraints and technical flaws in ML modeling even though their performance accuracy ranged from 0.65 to 1.00. While the development and internal validation were described for all models included (n=137), only 4.4% (6/137) have been validated in independent cohorts and 0.7% (1/137) have been assessed for clinical impact and efficacy.

Conclusion

Overall, the application of ML for modeling cancer outcomes in LLMICs is increasing. However, model development is largely unsatisfactory. We recommend model retraining using larger sample sizes, intensified external validation practices, and increased impact assessment studies using randomized controlled trial designs.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=308345, identifier CRD42022308345.

Development of a deep learning method for improving diagnostic accuracy for uterine sarcoma cases

Uterine sarcomas have very poor prognoses and are sometimes difficult to distinguish from uterine leiomyomas on preoperative examinations. Herein, we investigated whether deep neural network (DNN) models can improve the accuracy of preoperative MRI-based diagnosis in patients with uterine sarcomas. Fifteen sequences of MRI for patients (uterine sarcoma group: n = 63; uterine leiomyoma: n = 200) were used to train the models. Six radiologists (three specialists, three practitioners) interpreted the same images for validation. The most important individual sequences for diagnosis were axial T2-weighted imaging (T2WI), sagittal T2WI, and diffusion-weighted imaging. These sequences also represented the most accurate combination (accuracy: 91.3%), achieving diagnostic ability comparable to that of specialists (accuracy: 88.3%) and superior to that of practitioners (accuracy: 80.1%). Moreover, radiologists' diagnostic accuracy improved when provided with DNN results (specialists: 89.6%; practitioners: 92.3%). Our DNN models are valuable to improve diagnostic accuracy, especially in filling the gap of clinical skills between interpreters. This method can be a universal model for the use of deep learning in the diagnostic imaging of rare tumors.

Advances in the Preoperative Identification of Uterine Sarcoma

Simple Summary

As a lethal malignant tumor, uterine sarcomas lack specific diagnostic criteria due to their similar presentation with uterine fibroids, clinicians are prone to make the wrong diagnosis or adopt incorrect treatment methods, which leads to rapid tumor progression and increased metastatic propensity. In recent years, with the improvement of medical level and awareness of uterine sarcoma, more and more studies have proposed new methods for preoperative differentiation of uterine sarcoma and uterine fibroids. This review outlines the up-to-date knowledge about preoperative differentiation of uterine sarcoma and uterine fibroids, including laboratory tests, imaging examinations, radiomics and machine learning-related methods, preoperative biopsy, integrated model and other relevant emerging technologies, and provides recommendations for future research.

Abstract

Uterine sarcomas are rare malignant tumors of the uterus with a high degree of malignancy. Their clinical manifestations, imaging examination findings, and laboratory test results overlap with those of uterine fibroids. No reliable diagnostic criteria can distinguish uterine sarcomas from other uterine tumors, and the final diagnosis is usually only made after surgery based on histopathological evaluation. Conservative or minimally invasive treatment of patients with uterine sarcomas misdiagnosed preoperatively as uterine fibroids will shorten patient survival. Herein, we will summarize recent advances in the preoperative diagnosis of uterine sarcomas, including epidemiology and clinical manifestations, laboratory tests, imaging examinations, radiomics and machine learning-related methods, preoperative biopsy, integrated model and other relevant emerging technologies.

Preoperative Differentiation of Uterine Leiomyomas and Leiomyosarcomas: Current Possibilities and Future Directions

The distinguishing of uterine leiomyosarcomas (ULMS) and uterine leiomyomas (ULM) before the operation and histopathological evaluation of tissue is one of the current challenges for clinicians and researchers. Recently, a few new and innovative methods have been developed. However, researchers are trying to create different scales analyzing available parameters and to combine them with imaging methods with the aim of ULMs and ULM preoperative differentiation ULMs and ULM. Moreover, it has been observed that the technology, meaning machine learning models and artificial intelligence (AI), is entering the world of medicine, including gynecology. Therefore, we can predict the diagnosis not only through symptoms, laboratory tests or imaging methods, but also, we can base it on AI. What is the best option to differentiate ULM and ULMS preoperatively? In our review, we focus on the possible methods to diagnose uterine lesions effectively, including clinical signs and symptoms, laboratory tests, imaging methods, molecular aspects, available scales, and AI. In addition, considering costs and availability, we list the most promising methods to be implemented and investigated on a larger scale.

Classification of malignant tumors by a non-sequential recurrent ensemble of deep neural network model

Many significant efforts have so far been made to classify malignant tumors by using various machine learning methods. Most of the studies have considered a particular tumor genre categorized according to its originating organ. This has enriched the domain-specific knowledge of malignant tumor prediction, we are devoid of an efficient model that may predict the stages of tumors irrespective of their origin. Thus, there is ample opportunity to study if a heterogeneous collection of tumor images can be classified according to their respective stages. The present research work has prepared a heterogeneous tumor dataset comprising eight different datasets from The Cancer Imaging Archives and classified them according to their respective stages, as suggested by the American Joint Committee on Cancer. The proposed model has been used for classifying 717 subjects comprising different imaging modalities and varied Tumor-Node-Metastasis stages. A new non-sequential deep hybrid model ensemble has been developed by exploiting branched and re-injected layers, followed by bidirectional recurrent layers to classify tumor images. Results have been compared with standard sequential deep learning models and notable recent studies. The training and validation accuracy along with the ROC-AUC scores have been found satisfactory over the existing models. No model or method in the literature could ever classify such a diversified mix of tumor images with such high accuracy. The proposed model may help radiologists by acting as an auxiliary decision support system and speed up the tumor diagnosis process.

Radiomics and Artificial Intelligence in Uterine Sarcomas: A Systematic Review

Background: Recently, artificial intelligence (AI) with computerized imaging analysis is attracting the attention of clinicians, in particular for its potential applications in improving cancer diagnosis. This review aims to investigate the contribution of radiomics and AI on the radiological preoperative assessment of patients with uterine sarcomas (USs). Methods: Our literature review involved a systematic search conducted in the last ten years about diagnosis, staging and treatments with radiomics and AI in USs. The protocol was drafted according to the systematic review and meta-analysis preferred reporting project (PRISMA-P) and was registered in the PROSPERO database (CRD42021253535). Results: The initial search identified 754 articles; of these, six papers responded to the characteristics required for the revision and were included in the final analysis. The predominant technique tested was magnetic resonance imaging. The analyzed studies revealed that even though sometimes complex models included AI-related algorithms, they are still too complex for translation into clinical practice. Furthermore, since these results are extracted by retrospective series and do not include external validations, currently it is hard to predict the chances of their application in different study groups. Conclusion: To date, insufficient evidence supports the benefit of radiomics in USs. Nevertheless, this field is promising but the quality of studies should be a priority in these new technologies.

Artificial intelligence in gynecologic cancers: Current status and future challenges - A systematic review

OBJECTIVE

Over the past years, the application of artificial intelligence (AI) in medicine has increased rapidly, especially in diagnostics, and in the near future, the role of AI in medicine will become progressively more important. In this study, we elucidated the state of AI research on gynecologic cancers.

METHODS

A search was conducted in three databases-PubMed, Web of Science, and Scopus-for research papers dated between January 2010 and December 2020. As keywords, we used "artificial intelligence," "deep learning," "machine learning," and "neural network," combined with "cervical cancer," "endometrial cancer," "uterine cancer," and "ovarian cancer." We excluded genomic and molecular research, as well as automated pap-smear diagnoses and digital colposcopy.

RESULTS

Of 1632 articles, 71 were eligible, including 34 on cervical cancer, 13 on endometrial cancer, three on uterine sarcoma, and 21 on ovarian cancer. A total of 35 studies (49%) used imaging data and 36 studies (51%) used value-based data as the input data. Magnetic resonance imaging (MRI), computed tomography (CT), ultrasound, cytology, and hysteroscopy data were used as imaging data, and the patients' backgrounds, blood examinations, tumor markers, and indices in pathological examination were used as value-based data. The targets of prediction were definitive diagnosis and prognostic outcome, including overall survival and lymph node metastasis. The size of the dataset was relatively small because 64 studies (90%) included less than 1000 cases, and the median size was 214 cases. The models were evaluated by accuracy scores, area under the receiver operating curve (AUC), and sensitivity/specificity. Owing to the heterogeneity, a quantitative synthesis was not appropriate in this review.

CONCLUSIONS

In gynecologic oncology, more studies have been conducted on cervical cancer than on ovarian and endometrial cancers. Prognoses were mainly used in the study of cervical cancer, whereas diagnoses were primarily used for studying ovarian cancer. The proficiency of the study design for endometrial cancer and uterine sarcoma was unclear because of the small number of studies conducted. The small size of the dataset and the lack of a dataset for external validation were indicated as the challenges of the studies.

New imaging modalities to distinguish rare uterine mesenchymal cancers from benign uterine lesions

PURPOSE OF REVIEW

Uterine sarcomas are rare and are often challenging to differentiate on imaging from benign mimics, such as leiomyoma. As functional MRI techniques have improved and new adjuncts, such as machine learning and texture analysis, are now being investigated, it is helpful to be aware of the current literature on imaging features that may sometimes allow for preoperative distinction.

RECENT FINDINGS

MRI, with both conventional and functional imaging, is the modality of choice for evaluating uterine mesenchymal tumors, especially in differentiating uterine leiomyosarcoma from leiomyoma through validated diagnostic algorithms. MRI is sometimes helpful in differentiating high-grade stromal sarcoma from low-grade stromal sarcoma or differentiating endometrial stromal sarcoma from endometrial carcinoma. However, imaging remains nonspecific for evaluating rarer neoplasms, such as uterine tumor resembling ovarian sex cord tumor or perivascular epithelioid cell tumor, primarily because of the small number and power of relevant studies.

SUMMARY

Through advances in MRI techniques and novel investigational imaging adjuncts, such as machine learning and texture analysis, imaging differentiation of malignant from benign uterine mesenchymal tumors has improved and could help reduce morbidity relating to misdiagnosis or diagnostic delays.

Artificial intelligence: A rapid case for advancement in the personalization of Gynaecology/Obstetric and Mental Health care

To evaluate and holistically treat the mental health sequelae and potential psychiatric comorbidities associated with obstetric and gynaecological conditions, it is important to optimize patient care, ensure efficient use of limited resources and improve health-economic models. Artificial intelligence applications could assist in achieving the above. The World Health Organization and global healthcare systems have already recognized the use of artificial intelligence technologies to address 'system gaps' and automate some of the more cumbersome tasks to optimize clinical services and reduce health inequalities. Currently, both mental health and obstetric and gynaecological services independently use artificial intelligence applications. Thus, suitable solutions are shared between mental health and obstetric and gynaecological clinical practices, independent of one another. Although, to address complexities with some patients who may have often interchanging sequelae with mental health and obstetric and gynaecological illnesses, 'holistically' developed artificial intelligence applications could be useful. Therefore, we present a rapid review to understand the currently available artificial intelligence applications and research into multi-morbid conditions, including clinical trial-based validations. Most artificial intelligence applications are intrinsically data-driven tools, and their validation in healthcare can be challenging as they require large-scale clinical trials. Furthermore, most artificial intelligence applications use rate-limiting mock data sets, which restrict their applicability to a clinical population. Some researchers may fail to recognize the randomness in the data generating processes in clinical care from a statistical perspective with a potentially minimal representation of a population, limiting their applicability within a real-world setting. However, novel, innovative trial designs could pave the way to generate better data sets that are generalizable to the entire global population. A collaboration between artificial intelligence and statistical models could be developed and deployed with algorithmic and domain interpretability to achieve this. In addition, acquiring big data sets is vital to ensure these artificial intelligence applications provide the highest accuracy within a real-world setting, especially when used as part of a clinical diagnosis or treatment.

A meta-analysis of Watson for Oncology in clinical application

Using the method of meta-analysis to systematically evaluate the consistency of treatment schemes between Watson for Oncology (WFO) and Multidisciplinary Team (MDT), and to provide references for the practical application of artificial intelligence clinical decision-support system in cancer treatment. We systematically searched articles about the clinical applications of Watson for Oncology in the databases and conducted meta-analysis using RevMan 5.3 software. A total of 9 studies were identified, including 2463 patients. When the MDT is consistent with WFO at the ‘Recommended’ or the ‘For consideration’ level, the overall concordance rate is 81.52%. Among them, breast cancer was the highest and gastric cancer was the lowest. The concordance rate in stage I–III cancer is higher than that in stage IV, but the result of lung cancer is opposite (P < 0.05).Similar results were obtained when MDT was only consistent with WFO at the "recommended" level. Moreover, the consistency of estrogen and progesterone receptor negative breast cancer patients, colorectal cancer patients under 70 years old or ECOG 0, and small cell lung cancer patients is higher than that of estrogen and progesterone positive breast cancer patients, colorectal cancer patients over 70 years old or ECOG 1–2, and non-small cell lung cancer patients, with statistical significance (P < 0.05). Treatment recommendations made by WFO and MDT were highly concordant for cancer cases examined, but this system still needs further improvement. Owing to relatively small sample size of the included studies, more well-designed, and large sample size studies are still needed.

A combined radiomics and clinical variables model for prediction of malignancy in T2 hyperintense uterine mesenchymal tumors on MRI

OBJECTIVE

This study aims to develop a machine learning model for prediction of malignancy in T2 hyperintense mesenchymal uterine tumors based on T2-weighted image (T2WI) features and clinical information.

METHODS

This retrospective study included 134 patients with T2 hyperintense uterine mesenchymal tumors (104 patients in training cohort and 30 in testing cohort). A total of 960 radiomics features were initially computed and extracted from each 3D segmented tumor depicting on T2WI. The support vector machine (SVM) classifier was applied to build computer-aided diagnosis (CAD) models by using selected clinical and radiomics features, respectively. Finally, an observer study was conducted by comparing with two radiologists to evaluate the diagnostic performance. The area under the receiver operating characteristic (ROC) curve (AUC) was computed to assess the performance of each model.

RESULTS

Comparing with the T2WI-based radiomics model (AUC: 0.76 ± 0.09) and the clinical model (AUC: 0.79 ± 0.09), the combined model significantly improved the AUC value to 0.91 ± 0.05 (p < 0.05). The clinical-radiomics combined model yielded equivalent or higher performance than two radiologists (AUC: 0.78 vs. 0.91, p = 0.03; 0.90 vs.0.91, p = 0.13). There was a significant difference between the AUC values of two radiologists (p < 0.05).

CONCLUSIONS

It is feasible to predict malignancy risk of T2 hyperintense uterine mesenchymal tumors by combining clinical variables and T2WI-based radiomics features. Machine learning-based classification model may be useful to assist radiologists in decision-making.

KEY POINTS

• Radiomics approach has the potential to distinguish between benign and malignant mesenchymal uterine tumors. • T2WI-based radiomics analysis combined with clinical variables performed well in predicting malignancy risk of T2 hyperintense uterine mesenchymal tumors. • Machine learning-based classification model may be useful to assist radiologists in characterization of a T2 hyperintense uterine mesenchymal tumor.

Progress of Artificial Intelligence in Gynecological Malignant Tumors

Artificial intelligence (AI) is a sort of new technical science which can simulate, extend and expand human intelligence by developing theories, methods and application systems. In the last five years, the application of AI in medical research has become a hot topic in modern science and technology. Gynecological malignant tumors involves a wide range of knowledge, and AI can play an important part in these aspects, such as medical image recognition, auxiliary diagnosis, drug research and development, treatment scheme formulation and other fields. The purpose of this paper is to describe the progress of AI in gynecological malignant tumors and discuss some problems in its application. It is believed that AI improves the efficiency of diagnosis, reduces the burden of clinicians, and improves the effect of treatment and prognosis. AI will play an irreplaceable role in the field of gynecological malignant oncology and will promote the development of medicine and further promote the transformation from traditional medicine to precision medicine and preventive medicine. However, there are also some problems in the application of AI in gynecologic malignant tumors. For example, AI, inseparable from human participation, still needs to be more “humanized”, and needs to further protect patients’ privacy and health, improve legal and insurance protection, and further improve according to local ethnic conditions and national conditions. However, it is believed that with the continuous development of AI, especially ensemble classifier, and deep learning will have a profound influence on the future of medical technology, which is a powerful driving force for future medical innovation and reform.

A Diagnostic Algorithm using Multi-parametric MRI to Differentiate Benign from Malignant Myometrial Tumors: Machine-Learning Method

This study aimed to develop a diagnostic algorithm for preoperative differentiating uterine sarcoma from leiomyoma through a supervised machine-learning method using multi-parametric MRI. A total of 65 participants with 105 myometrial tumors were included: 84 benign and 21 malignant lesions (belonged to 51 and 14 patients, respectively; based on their postoperative tissue diagnosis). Multi-parametric MRI including T1-, T2-, and diffusion-weighted (DW) sequences with ADC-map, contrast-enhanced images, as well as MR spectroscopy (MRS), was performed for each lesion. Thirteen singular MRI features were extracted from the mentioned sequences. Various combination sets of selective features were fed into a machine classifier (coarse decision-tree) to predict malignant or benign tumors. The accuracy metrics of either singular or combinational models were assessed. Eventually, two diagnostic algorithms, a simple decision-tree and a complex one were proposed using the most accurate models. Our final simple decision-tree obtained accuracy = 96.2%, sensitivity = 100% and specificity = 95%; while the complex tree yielded accuracy, sensitivity and specificity of 100%. To summarise, the complex diagnostic algorithm, compared to the simple one, can differentiate tumors with equal sensitivity, but a higher specificity and accuracy. However, it needs some further time-consuming modalities and difficult imaging calculations. Trading-off costs and benefits in appropriate situations must be determinative.

Investigating the diagnostic value of quantitative parameters based on T2-weighted and contrast-enhanced MRI with psoas muscle and outer myometrium as internal references for differentiating uterine sarcomas from leiomyomas at 3T MRI

Background

Post-hysterectomy histopathological examination is currently the main diagnostic tool for differentiating uterine sarcomas from leiomyomas. This study aimed to investigate the diagnostic accuracy of preoperative quantitative metrics based on T2-weighted sequences and contrast-enhanced MRI (CE-MRI) for distinguishing uterine sarcomas from leiomyomas.

Materials and methods

The institutional review board approved the study. Sixty-five women confirmed to have a total of 105 lesions participated. Routine pelvic MRI sequences, T2 map and CE-MRI images were performed preoperatively using a 3 T MR scanner. Six quantitative metrics—T2 mapping parameter, T2 scaled ratio, tumor myometrium contrast ratio on T2, tumor psoas contrast ratio on T2, tumor myometrium contrast-enhanced ratio, and tumor psoas contrast-enhanced ratio—were extracted from the acquired image sets. Chi-square test was used to compare the percentage of malignant lesions with the central necrosis to the corresponding percentage for the benign masses. Using the area under receiver operating characteristic (AUC) curve, the performance of different metrics for distinguishing uterine sarcomas from leiomyomas was measured. Moreover, for each metric, we extracted the optimal cut-off value. The values of sensitivity, specificity, negative predictive value, and positive predictive value were calculted for the classifiers based on different metrics.

Results

The average age, average lesion size, and proportion of premenopausal women in benign and malignant groups were comparable in our dataset. The signal intensity of uterine sarcomas at T2-weighted sequences was significantly higher than that of leiomyomas (p < 0.001), while intensity at T1-weighted sequences exhibited no significant difference between the two masses (p = 0.201). Our data also suggested that a central necrosis was ten times more common among malignant lesions compared to benign ones (p < 0.001). Among different metrics, T2 mapping parameter achieved the highest AUC value and accuracy in differentiating two groups. Three measures—T2 scaled ratio, tumor myometrium contrast ratio on T2, and tumor myometrium contrast-enhanced ratio—achieved a sensitivity of 100%, therefore none of the malignant lesions would have been missed if these metrics had been adopted in patient management.

Conclusions

The findings suggested that the evaluated metrics could be useful in the preoperative assessment of myometrial masses to differentiate uterine sarcomas from leiomyomas. The proposed framework has major implications for improving current practice in the management of myometrial masses.