Emily Skelton

Bio: Emily Skelton is an academic researcher from King's College London. The author has contributed to research in topics: Medicine & Segmentation. The author has an hindex of 7, co-authored 39 publications receiving 224 citations. Previous affiliations of Emily Skelton include City University London & East Sussex County Council.

Image-guided core needle biopsy in the diagnosis of malignant lymphoma

Abstract: Objective Current European Society for Medical Oncology (ESMO) guidelines recommend that when feasible, surgical excision biopsy (SEB) is the ideal for diagnosis, sub-typing and grading of malignant lymphoma. We undertook this retrospective study to assess the diagnostic accuracy of image-guided core needle biopsy (CNB) in the diagnosis of malignant lymphoma, to identify the proportion of cases from which oncological treatment was subsequently instigated from the CNB diagnosis, and to evaluate the potential role for minimally invasive CNB techniques in the diagnostic pathway of malignant lymphoma. Methods All cases of lymphoma amenable to CNB between 2008 and 2013 were included. Patient records were reviewed to identify the biopsy diagnostic pathway undertaken (fine needle aspiration cytology, CNB, surgical excision biopsy). CNB specimens were graded as fully diagnostic (tumour sub-typing/grading and treatment initiated), partially diagnostic (diagnosis of lymphoma but more tissue required for sub-typing/grading), equivocal or inadequate. The effects of anatomical location, needle gauge, number of core specimens and sub-type of disease on the diagnostic yield of the sample were analysed. Results 262 patients and 323 biopsy specimens were included in the study. 237 patients underwent CNB as the initial diagnostic intervention. In 230/237 CNB was fully diagnostic (97%), allowing initiation of treatment. In 7 patients, SEB was necessary in addition to CNB to provide additional diagnostic information to allow initiation of treatment. In 72 patients, SEB was the only diagnostic test performed. Conclusion Our study showed that in 97% of suitable cases, CNB provided sufficient diagnostic information to allow treatment of malignant lymphoma to be instigated. This minimally-invasive technique is well tolerated and has advantages over surgical techniques, including reduced costs, post-procedural complications and delays on the diagnostic pathway. CNB may obviate the use of surgical techniques in the majority of suitable cases, however its success is dependent on close collaboration and acceptance by clinicians and pathologists.

Establishing an accurate diagnosis of a parotid lump: evaluation of the current biopsy methods - fine needle aspiration cytology, ultrasound-guided core biopsy, and intraoperative frozen section.

Abstract: The optimum technique for histological confirmation of the nature of a parotid mass remains controversial. Fine needle aspiration cytology (FNAC), which has traditionally been used, is associated with high non-diagnostic and false negative rates, and ultrasound (US)-guided core biopsy and frozen section have been explored as alternatives. US-guided core biopsy is more invasive than FNAC, but is safe, well-tolerated, and associated with improved diagnostic performance. Although frozen section offers better specificity than FNAC, it has a number of important drawbacks and cannot be considered as a primary diagnostic tool. US-guided core biopsy should be considered as the initial diagnostic technique of choice, and in units where the accuracy of FNAC is good it can be used when FNAC is equivocal or non-diagnostic.

LSTM Spatial Co-transformer Networks for Registration of 3D Fetal US and MR Brain Images

Abstract: In this work, we propose a deep learning-based method for iterative registration of fetal brain images acquired by ultrasound and magnetic resonance, inspired by “Spatial Transformer Networks”. Images are co-aligned to a dual modality spatio-temporal atlas, where computational image analysis may be performed in the future. Our results show better alignment accuracy compared to “Self-Similarity Context descriptors”, a state-of-the-art method developed for multi-modal image registration. Furthermore, our method is robust and able to register highly misaligned images, with any initial orientation, where similarity-based methods typically fail.

Deep learning in medical image registration: a survey

Abstract: The establishment of image correspondence through robust image registration is critical to many clinical tasks such as image fusion, organ atlas creation, and tumor growth monitoring and is a very challenging problem. Since the beginning of the recent deep learning renaissance, the medical imaging research community has developed deep learning-based approaches and achieved the state-of-the-art in many applications, including image registration. The rapid adoption of deep learning for image registration applications over the past few years necessitates a comprehensive summary and outlook, which is the main scope of this survey. This requires placing a focus on the different research areas as well as highlighting challenges that practitioners face. This survey, therefore, outlines the evolution of deep learning-based medical image registration in the context of both research challenges and relevant innovations in the past few years. Further, this survey highlights future research directions to show how this field may be possibly moved forward to the next level.

Deep learning in medical image registration: a review.

Abstract: This paper presents a review of deep learning (DL)-based medical image registration methods. We summarized the latest developments and applications of DL-based registration methods in the medical field. These methods were classified into seven categories according to their methods, functions and popularity. A detailed review of each category was presented, highlighting important contributions and identifying specific challenges. A short assessment was presented following the detailed review of each category to summarize its achievements and future potential. We provided a comprehensive comparison among DL-based methods for lung and brain registration using benchmark datasets. Lastly, we analyzed the statistics of all the cited works from various aspects, revealing the popularity and future trend of DL-based medical image registration.

Deep Learning in Medical Image Registration: A Survey

Abstract: The establishment of image correspondence through robust image registration is critical to many clinical tasks such as image fusion, organ atlas creation, and tumor growth monitoring, and is a very challenging problem. Since the beginning of the recent deep learning renaissance, the medical imaging research community has developed deep learning based approaches and achieved the state-of-the-art in many applications, including image registration. The rapid adoption of deep learning for image registration applications over the past few years necessitates a comprehensive summary and outlook, which is the main scope of this survey. This requires placing a focus on the different research areas as well as highlighting challenges that practitioners face. This survey, therefore, outlines the evolution of deep learning based medical image registration in the context of both research challenges and relevant innovations in the past few years. Further, this survey highlights future research directions to show how this field may be possibly moved forward to the next level.