Non-Invasive Radiomics Approach Predict Invasiveness of Adamantinomatous Craniopharyngioma Before Surgery

Machine learning models from computed tomography to diagnose thymic epithelial tumors requiring combined resection

Background

Minimally invasive approaches have been a standard choice of surgery for noninvasive thymic epithelial tumors (TETs), but we sometimes experience cases requiring combined resection of adjacent structures. We develop and validate machine learning models to predict combined resection based on preoperative contrast-enhanced computed tomography (CT).

Methods

This study included 212 patients with TETs (140 in the training cohort and 72 in the validation cohort) who underwent radical surgery. Radiomics features were extracted from contrast-enhanced CT and predicted with five feature selection methods and seven machine learning models in nested cross validation. The clinical utility of the models was analyzed by a decision curve analysis (DCA).

Results

Fifty-five patients in the training cohort and 28 in the validation cohort required combined resection. The classifiers random forest (RF), gradient boosting (GB), and eXtreme Gradient Boosting (XGB) indicated high predictive performance, with the XGB classifier based on features selected by GB performing the best, with an area under the curve (AUC) of 0.797. In the validation cohort, the classifier had an AUC of 0.817. The DCA showed the validity of the model with a threshold range of 15-72%. When restricted to combined pulmonary and pericardial resection, the respective AUCs were 0.736 and 0.674 for the training cohort and 0.806 and 0.924 for the validation cohort.

Conclusions

The machine learning model based on preoperative CT images was able to diagnose TETs requiring combined resection with high accuracy. The DCA demonstrated a wide range of model validity and may aid in surgical approach selection.

Topography and Radiological Variables as Ancillary Parameters for Evaluating Tissue Adherence, Hypothalamic-Pituitary Dysfunction, and Recurrence in Craniopharyngioma: An Integrated Multidisciplinary Overview

Craniopharyngiomas continue to present a challenge in clinical practice due to their heterogeneity and unpredictable adherence to vital neurovascular structures, particularly the hypothalamus. This results in different degrees of hypothalamus-pituitary axis dysfunction and a lack of uniform consensus and treatment guidelines regarding optimal management. MRI and CT are complementary techniques in the preoperative diagnostic phase, enabling the precise definition of craniopharyngioma size, shape, and consistency, as well as guiding classification into histopathological subtypes and topographical categories. Meanwhile, MRI plays a crucial role in the immediate postoperative period and follow-up stages by identifying treatment-related changes and residual tumors. This pictorial essay aims to provide an overview of the role of imaging in identifying variables indicative of the adherence degree to the hypothalamus, hypothalamic-pituitary dysfunction, the extent of surgical excision, and prognosis. For a more comprehensive assessment, we choose to distinguish the following two scenarios: (1) the initial diagnosis phase, where we primarily discuss the role of radiological variables predictive of adhesions to the surrounding neurovascular structures and axis dysfunction which may influence the choice of surgical resection; (2) the early post-treatment follow-up phase, where we discuss the interpretation of treatment-related changes that impact outcomes.

Advances in the treatment of Adamantinomatous craniopharyngioma: How to balance tumor control and quality of life in the current environment: a narrative review

Adamantinomatous craniopharyngioma (ACP) presents a significant challenge to neurosurgeons despite its benign histology due to its aggressive behavior and unique growth patterns. This narrative review explores the evolving landscape of ACP treatments and their efficacy, highlighting the continuous development in therapeutic approaches in recent years. Traditionally, complete resection was the primary treatment for ACP, but surgical -related morbidity have led to a shift. The invasive nature of the finger-like protrusions in the histological structure results in a higher recurrence rate for ACP compared to papillary craniopharyngioma (PCP), even after complete macroscopic resection. Given this, combining subtotal resection with adjuvant radiotherapy has shown potential for achieving similar tumor control rates and potentially positive endocrine effects. Simultaneously, adjuvant treatments (such as radiotherapy, intracystic treatment, and catheter implantation) following limited surgery offer alternative approaches for sustained disease control while minimizing morbidity and alleviating clinical symptoms. Additionally, advances in understanding the molecular pathways of ACP have paved the way for targeted drugs, showing promise for therapy. There is a diversity of treatment models for ACP, and determining the optimal approach remains a subject of ongoing debate in the present context. In order to achieve a good-term quality of life (QOL), the main goal of the cyst disappearance or reduction of surgical treatment is still the main. Additionally, there should be a greater emphasis on personalized treatment at this particular stage and the consideration of ACP as a potentially chronic neurosurgical condition. This review navigates the evolving landscape of ACP therapies, fostering ongoing discussions in this complex field.

Contemporary Biological Insights and Clinical Management of Craniopharyngioma

Craniopharyngiomas (CPs) are clinically aggressive tumors because of their invasive behavior and recalcitrant tendency to recur after therapy. There are 2 types based on their distinct histology and molecular features: the papillary craniopharyngioma (PCP), which is associated with BRAF-V600E mutations and the adamantinomatous craniopharyngioma (ACP), characterized by mutations in CTNNB1 (encoding β-catenin). Patients with craniopharyngioma show symptoms linked to the location of the tumor close to the optic pathways, hypothalamus, and pituitary gland, such as increased intracranial pressure, endocrine deficiencies, and visual defects. Treatment is not specific and mostly noncurative, and frequently includes surgery, which may achieve gross total or partial resection, followed by radiotherapy. In cystic tumors, frequent drainage is often required and intracystic instillation of drugs has been used to help manage cyst refilling. More recently targeted therapies have been used, particularly in PCP, but also now in ACP and clinical trials are underway or in development. Although patient survival is high, the consequences of the tumor and its treatment can lead to severe comorbidities resulting in poor quality of life, in particular for those patients who bear tumors with hypothalamic involvement. Accordingly, in these patients at risk for the development of a hypothalamic syndrome, hypothalamus-sparing treatment strategies such as limited resection followed by irradiation are recommended. In this review, we provide an update on various aspects of CP, with emphasis on recent advances in the understanding of tumor pathogenesis, clinical consequences, management, and therapies.

Automated segmentation of craniopharyngioma on MR images using U-Net-based deep convolutional neural network

OBJECTIVES

To develop a U-Net-based deep learning model for automated segmentation of craniopharyngioma.

METHODS

A total number of 264 patients diagnosed with craniopharyngiomas were included in this research. Pre-treatment MRIs were collected, annotated, and used as ground truth to learn and evaluate the deep learning model. Thirty-eight patients from another institution were used for independently external testing. The proposed segmentation model was constructed based on a U-Net architecture. Dice similarity coefficients (DSCs), Hausdorff distance of 95% percentile (95HD), Jaccard value, true positive rate (TPR), and false positive rate (FPR) of each case were calculated. One-way ANOVA analysis was used to investigate if the model performance was associated with the radiological characteristics of tumors.

RESULTS

The proposed model showed a good performance in segmentation with average DSCs of 0.840, Jaccard of 0.734, TPR of 0.820, FPR of 0.000, and 95HD of 3.669 mm. It performed feasibly in the independent external test set, with average DSCs of 0.816, Jaccard of 0.704, TPR of 0.765, FPR of 0.000, and 95HD of 4.201 mm. Also, one-way ANOVA suggested the performance was not statistically associated with radiological characteristics, including predominantly composition (p = 0.370), lobulated shape (p = 0.353), compressed or enclosed ICA (p = 0.809), and cavernous sinus invasion (p = 0.283).

CONCLUSIONS

The proposed deep learning model shows promising results for the automated segmentation of craniopharyngioma.

KEY POINTS

• The segmentation model based on U-Net showed good performance in segmentation of craniopharyngioma. • The proposed model showed good performance regardless of the radiological characteristics of craniopharyngioma. • The model achieved feasibility in the independent external dataset obtained from another center.

Craniopharyngioma surgery for rhinologists

PURPOSE OF REVIEW

Craniopharyngiomas are rare tumors that that present with loss of pituitary function in most cases. They present in a bimodal age distribution and the most common treatment paradigms include gross total resection or subtotal resection followed by radiation. Endoscopic transnasal access to intradural tumors has become increasingly common due to improvements in equipment, increased familiarity with the surgical corridor and anatomy, and reconstruction techniques. As such, rhinologists play an increasingly important role in the management of craniopharyngiomas.

RECENT FINDINGS

Recent years have highlighted our growing experience with pediatric endonasal skull base surgery. Prior concerns including sphenoid pneumatization, midfacial growth restrictions, and intercarotid space limitations have been studied more extensively. It has been found that there are no increased complications with lack of sphenoid pneumatization, no changes to midfacial growth with endonasal techniques, and the inter-carotid distance is stable after around age 5. Advances in surgical and skull base reconstruction techniques and intraoperative monitoring have reduced the risks of complications from surgery.

SUMMARY

Rhinologists play an important role in craniopharyngioma surgery. The approach to and reconstruction after tumor removal are vital portions of the procedure that allow for resection and prevent postsurgical complications.

Pathological Diagnosis of Adult Craniopharyngioma on MR Images: An Automated End-to-End Approach Based on Deep Neural Networks Requiring No Manual Segmentation

PURPOSE

The goal of this study was to develop end-to-end convolutional neural network (CNN) models that can noninvasively discriminate papillary craniopharyngioma (PCP) from adamantinomatous craniopharyngioma (ACP) on MR images requiring no manual segmentation.

MATERIALS AND METHODS

A total of 97 patients diagnosed with ACP or PCP were included. Pretreatment contrast-enhanced T1-weighted images were collected and used as the input of the CNNs. Six models were established based on six networks, including VGG16, ResNet18, ResNet50, ResNet101, DenseNet121, and DenseNet169. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity were used to assess the performances of these deep neural networks. A five-fold cross-validation was applied to evaluate the performances of the models.

RESULTS

The six networks yielded feasible performances, with area under the receiver operating characteristic curves (AUCs) of at least 0.78 for classification. The model based on Resnet50 achieved the highest AUC of 0.838 ± 0.062, with an accuracy of 0.757 ± 0.052, a sensitivity of 0.608 ± 0.198, and a specificity of 0.845 ± 0.034, respectively. Moreover, the results also indicated that the CNN method had a competitive performance compared to the radiomics-based method, which required manual segmentation for feature extraction and further feature selection.

CONCLUSIONS

MRI-based deep neural networks can noninvasively differentiate ACP from PCP to facilitate the personalized assessment of craniopharyngiomas.

Machine Learning for the Detection and Segmentation of Benign Tumors of the Central Nervous System: A Systematic Review

Simple Summary

Machine learning in radiology of the central nervous system has seen many interesting publications in the past few years. Since the focus has largely been on malignant tumors such as brain metastases and high-grade gliomas, we conducted a systematic review on benign tumors to summarize what has been published and where there might be gaps in the research. We found several studies that report good results, but the descriptions of methodologies could be improved to enable better comparisons and assessment of biases.

Abstract

Objectives: To summarize the available literature on using machine learning (ML) for the detection and segmentation of benign tumors of the central nervous system (CNS) and to assess the adherence of published ML/diagnostic accuracy studies to best practice. Methods: The MEDLINE database was searched for the use of ML in patients with any benign tumor of the CNS, and the records were screened according to PRISMA guidelines. Results: Eleven retrospective studies focusing on meningioma (n = 4), vestibular schwannoma (n = 4), pituitary adenoma (n = 2) and spinal schwannoma (n = 1) were included. The majority of studies attempted segmentation. Links to repositories containing code were provided in two manuscripts, and no manuscripts shared imaging data. Only one study used an external test set, which raises the question as to whether some of the good performances that have been reported were caused by overfitting and may not generalize to data from other institutions. Conclusions: Using ML for detecting and segmenting benign brain tumors is still in its infancy. Stronger adherence to ML best practices could facilitate easier comparisons between studies and contribute to the development of models that are more likely to one day be used in clinical practice.

Nomograms to Predict Endocrinological Deficiency in Patients With Surgically Treated Craniopharyngioma

Objective

Postoperative hypopituitarism associated with increased risks of premature mobility and mortality is often encountered in craniopharyngioma patients. The aim of our study is to construct nomograms related to injury types of the hypothalamus-pituitary axis (HPA) to predict hypopituitarism 1 year after surgery.

Methods

Craniopharyngioma patients undergoing initial endoscopic endonasal surgery between December 2012 and March 2021 in our center were retrospectively reviewed, and injury types of the HPA were categorized according to intraoperative endoscopic observation. Included patients were randomly divided into a training group and a validation group. Nomograms were established based on the results of multivariate logistic analysis. The predictive performance of the nomograms was evaluated in the training and validation groups.

Results

A total of 183 patients with craniopharyngioma were enrolled, and seven injury types of the HPA were summarized. Relative to intact HPA, exclusive hypothalamus injury significantly increased the risk of anterior (OR, 194.174; 95% CI, 21.311-1769.253; p < 0.001) and posterior pituitary dysfunction (OR, 31.393; 95% CI, 6.319-155.964; p < 0.001) 1 year after surgery, while exclusively sacrificing stalk infiltrated by tumors did not significantly increase the risk of anterior (OR, 5.633; 95% CI, 0.753-42.133; p = 0.092) and posterior pituitary dysfunction (OR, 1.580; 95% CI, 0.257-9.707; p = 0.621) 1 year after surgery. In the training group, the AUCs of nomograms predicting anterior and posterior pituitary dysfunction 1 year after surgery were 0.921 and 0.885, respectively, compared with 0.921 and 0.880 in the validation group.

Conclusions

Intact hypothalamus structure is critical in maintaining pituitary function. Moreover, our preliminary study suggests that the pituitary stalk infiltrated by craniopharyngioma could be sacrificed to achieve radical resection, without substantially rendering significantly worse endocrinological efficiency 1 year after surgery. The user-friendly nomograms can be used to predict hypopituitarism 1 year after surgery.

A multi-task two-path deep learning system for predicting the invasiveness of craniopharyngioma

BACKGROUND AND OBJECTIVE

Craniopharyngioma is a kind of benign brain tumor in histography. However, it might be clinically aggressive and have severe manifestations, such as increased intracranial pressure, hypothalamic-pituitary dysfunction, and visual impairment. It is considered challenging for radiologists to predict the invasiveness of craniopharyngioma through MRI images. Therefore, developing a non-invasive method that can predict the invasiveness and boundary of CP as a reference before surgery is of clinical value for making more appropriate and individualized treatment decisions and reducing the occurrence of inappropriate surgical plan choices.

METHODS

The MT-Brain system has consisted of two pathways, a sub-path based on 2D CNN for capturing the features from each slice of MRI images, and a 3D sub-network for capturing additional context information between slices. By introducing the two-path architecture, our system can make full use of the fusion of the above 2D and 3D features for classification. Furthermore, position encoding and mask-guided attention also have been introduced to improve the segmentation and diagnosis performance. To verify the performance of the MT-Brain system, we have enrolled 1032 patients with craniopharyngioma (302 invasion and 730 non-invasion patients), segmented the tumors on postcontrast coronal T1WI and randomized them into a training dataset and a testing dataset at a ratio of 8:2.

RESULTS

The MT-Brain system achieved a remarkable performance in diagnosing the invasiveness of craniopharyngioma with the AUC of 83.84%, the accuracy of 77.94%, the sensitivity of 70.97%, and the specificity of 80.99%. In the lesion segmentation task, the predicted boundaries of lesions were similar to those labeled by radiologists with the dice of 66.36%. In addition, some explorations also have been made on the interpretability of deep learning models, illustrating the reliability of the model.

CONCLUSIONS

To the best of our knowledge, this study is the first to develop an integrated deep learning model to predict the invasiveness of craniopharyngioma preoperatively and locate the lesion boundary synchronously on MRI. The excellent performances indicate that the MT-Brain system has great potential in real-world clinical applications.

Current Advances in the Management of Adult Craniopharyngiomas

Craniopharyngiomas (CPs) are slow growing, histologically benign intracranial tumors located in the sellar–suprasellar region. Although known to have low mortality, their location and relationship to the adjacent neural structures results in patients having significant neurologic, endocrine, and visual comorbidities. The invasive nature of this tumor makes complete resection a challenge and contributes to its recurrence. Additionally, these tumors are bimodally distributed, being treated with surgery, and are followed by other adjuncts, such as focused radiation therapy, e.g., Gamma knife. Advances in surgical techniques, imaging tools, and instrumentations have resulted in the evolution of surgery using endoscopic techniques, with residual components being treated by radiotherapy to target the residual tumor. Advances in molecular biology have elucidated the main pathways involved in tumor development and recurrence, but presently, no other treatments are offered to patients, besides surgery, radiation, and endocrine management, as the disease and tumor evolve. We review the contemporary management of these tumors, from the evolution of surgical treatments, utilizing standard open microscopic approaches to the more recent endoscopic surgery, and discuss the current recommendations for care of these patients. We discuss the developments in radiation therapy, such as radiosurgery, being used as treatment strategies for craniopharyngioma, highlighting their beneficial effects on tumor resections while decreasing the rates of adverse outcomes. We also outline the recent chemotherapy modalities, which help control tumor growth, and the immune landscape on craniopharyngiomas that allow the development of novel immunotherapies.

Clinical Outcomes of Transcranial and Endoscopic Endonasal Surgery for Craniopharyngiomas: A Single-Institution Experience

OBJECTIVE

Craniopharyngioma has always been a challenge for the neurosurgeon, and there is no consensus on optimal treatment. The objective of this study was to compare surgical outcomes and complications between transcranial surgery (TCS) and endoscopic endonasal surgery (EES) of craniopharyngiomas.

METHODS

A retrospective review of patients who underwent craniopharyngioma resection at Wuhan Union Hospital between January 2010 and December 2019 was performed. A total of 273 patients were enrolled in this retrospective study. All patients were analyzed with surgical effects, endocrinologic outcomes, complications, and follow-up results.

RESULTS

A total of 185 patients underwent TCS and 88 underwent EES. There were no significant differences in patient demographic data, preoperative symptoms, and tumor characteristics between the two groups. The mean follow-up was 30.5 months (range 8-51 months). The EES group had a greater gross total resection (GTR) rate (89.8% EES vs. 77.3% TCS, p < 0.05) and lower rate of hypopituitarism (53.4% EES vs. 68.1% TCS, p < 0.05) and diabetes insipidus (DI) (51.1% EES vs. 72.4% TCS, p < 0.05). More postoperative cerebrospinal fluid (CSF) leaks occurred in the EES group (4.5% EES vs. 0% TCS, p < 0.05). More patients in the EES group with preoperative visual deficits experienced improvement after surgery (74.5% EES vs. 56.3% TCS, p < 0.05). There were statistical differences in the recurrence rates (12.5% EES vs. 23.8% TCS, p < 0.05) between the 2 groups.

CONCLUSION

These data support the view that EES is a safe and effective minimally invasive surgery compared to TCS. Compared to TCS, EES has fewer surgical complications and a lower recurrence rate.

Application of Artificial Intelligence in Diagnosis of Craniopharyngioma

Craniopharyngioma is a congenital brain tumor with clinical characteristics of hypothalamic-pituitary dysfunction, increased intracranial pressure, and visual field disorder, among other injuries. Its clinical diagnosis mainly depends on radiological examinations (such as Computed Tomography, Magnetic Resonance Imaging). However, assessing numerous radiological images manually is a challenging task, and the experience of doctors has a great influence on the diagnosis result. The development of artificial intelligence has brought about a great transformation in the clinical diagnosis of craniopharyngioma. This study reviewed the application of artificial intelligence technology in the clinical diagnosis of craniopharyngioma from the aspects of differential classification, prediction of tissue invasion and gene mutation, prognosis prediction, and so on. Based on the reviews, the technical route of intelligent diagnosis based on the traditional machine learning model and deep learning model were further proposed. Additionally, in terms of the limitations and possibilities of the development of artificial intelligence in craniopharyngioma diagnosis, this study discussed the attentions required in future research, including few-shot learning, imbalanced data set, semi-supervised models, and multi-omics fusion.

Application of artificial intelligence and radiomics in pituitary neuroendocrine and sellar tumors: a quantitative and qualitative synthesis

PURPOSE

To systematically review the literature regarding the application of machine learning (ML) of magnetic resonance imaging (MRI) radiomics in common sellar tumors. To identify future directions for application of ML in sellar tumor MRI.

METHODS

PubMed, Medline, Embase, Google Scholar, Scopus, ArxIV, and bioRxiv were searched to identify relevant studies published between 2010 and September 2021. Studies were included if they specifically involved ML of MRI radiomics in the analysis of sellar masses. Risk of bias assessment was performed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) Tool.

RESULTS

Fifty-eight articles were identified for review. All papers utilized retrospective data, and a quantitative systematic review was performed for thirty-one studies utilizing a public dataset which compared pituitary adenomas, meningiomas, and gliomas. One of the analyzed architectures yielded the highest classification accuracy of 0.996. The remaining twenty-seven articles were qualitatively reviewed and showed promising findings in predicting specific tumor characteristics such as tumor consistency, Ki-67 proliferative index, and post-surgical recurrence.

CONCLUSION

This review highlights the potential clinical application of ML using MRI radiomic data of the sellar region in diagnosis and predicting treatment outcomes. We describe future directions for practical application in the clinical care of patients with pituitary neuroendocrine and other sellar tumors.