Archives

  • 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • A Haegens J H Vernooy P Heeringa B

    2020-08-18

    [46] A. Haegens, J.H. Vernooy, P. Heeringa, B.T. Mossman, E.F. Wouters, Myeloperoxidase modulates lung epithelial responses to pro-inflammatory agents, Eur. Respir. J. 31 (2008) 252–260. [47] R. Lande, D. Ganguly, V. Facchinetti, L. Frasca, C. Conrad, J. Gregorio, S. Meller, G. Chamilos, R. Sebasigari, V. Riccieri, R. Bassett, H. Amuro, S. Fukuhara, T. Ito, Y.J. Liu, M. Gilliet, Neutrophils activate plasmacytoid dendritic FG-4592 by releasing self-DNA-peptide complexes in systemic lupus erythematosus, Sci. Transl. Med. 3 (2011) 73ra19.
    Contents lists available at ScienceDirect
    Radiotherapy and Oncology
    Original Article
    Benefits of deep learning for delineation of organs at risk in head and neck cancer
    J. van der Veen a,1, S. Willems b,1, S. Deschuymer a, D. Robben b,c, W. Crijns a, F. Maes b, S. Nuyts a,⇑ a KU Leuven, Dept. Oncology, Laboratory of Experimental Radiotherapy, & UZ Leuven, Radiation Oncology, Belgium; b KU Leuven, Dept. ESAT, Processing Speech and Images (PSI), & UZ Leuven, Medical Imaging Research Center, Belgium; c Icometrix, B-3000 Leuven, Belgium
    Article history:
    Keywords:
    Head and neck neoplasms
    Radiotherapy
    Organs at risk
    Observer variation
    Neural networks (computer)
    Delineation 
    Purpose/objective: Precise delineation of organs at risk (OARs) in head and neck cancer (HNC) is necessary for accurate radiotherapy. Although guidelines exist, significant interobserver variability (IOV) remains. The aim was to validate a 3D convolutional neural network (CNN) for semi-automated delineation of OARs with respect to delineation accuracy, efficiency and consistency compared to manual delineation. Material/methods: 16 OARs were manually delineated in 15 new HNC patients by two trained radiation oncologists (RO) independently, using international consensus guidelines. OARs were also automatically delineated by applying the CNN and corrected as needed by both ROs separately. Both delineations were performed two weeks apart and blinded to each other. IOV between both ROs was quantified using Dice similarity coefficient (DSC) and average symmetric surface distance (ASSD). To objectify network accu-racy, differences between automated and corrected delineations were calculated using the same similar-ity measures.
    Results: Average correction time of the automated delineation was 33% shorter than manual delineation (23 vs 34 minutes) (p < 10–6). IOV improved significantly with network initialisation for nearly all OARs (p < 0.05), resulting in decreased ASSD averaged over all OARs from 1.9 to 1.2 mm. The network achieved an accuracy of 90% and 84% DSC averaged over all OARs for RO1 and RO2 respectively, with an ASSD of 0.7 and 1.5 mm, which was in 93% and 73% of the cases lower than the IOV.
    Conclusion: The CNN developed for automated OAR delineation in HNC was shown to be more efficient and consistent compared to manual delineation, which justify its implementation in clinical practice.
    2019 Elsevier B.V. All rights reserved. Radiotherapy and Oncology xxx (2019) xxx–xxx
    Ranked as the seventh most common cancer and cause of can-cer death worldwide, the burden of head and neck cancer (HNC) on society and the health sector should not be underestimated [1]. Radiotherapy (RT) plays an important role in the curative treat-ment of HNC and allows organ preservation and improved function preservation in selected cases, compared to surgery. Intensification of RT regimens by means of altered fractionation and concomitant chemotherapy have been beneficial for overall survival and loco-regional control [2,3], although disease recurrence remains an issue [4,5]. At the same time, intensification of RT regimens has induced an increase in acute and late toxicity, limiting further
    Abbreviations: ASSD, average symmetric surface distance; mm, millimetres; Acc, accuracy of network; RO, radiation oncologist; PCM, pharyngeal constrictor muscles; PG, parotid gland; SG, submandibular gland; U, upper; S, supra; IOVm, manual interobserver variability. ⇑ Corresponding author at: KU Leuven, Dept. Oncology, Laboratory of Experi-mental Radiotherapy, & UZ Leuven, Radiation Oncology, B-3000 Leuven, Belgium.
    E-mail address: [email protected] (S. Nuyts).
    1 These authors contributed equally to this work.
    treatment intensification [6]. To compensate for this, implementa-tion of intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy (VMAT) have allowed the dose delivered to the tumour to be shaped, resulting in better sparing of normal and critical tissue, decreasing toxicity [7]. Proton therapy has the potential to spare organs at risk (OARs) even more due to its more favourable dose-depth characteristics, including a sharp localised high dose delivery at the Bragg peak, with a low exit dose [8]. Adaptive radiotherapy, with the intention of sparing normal tissue even better and still provide sufficient coverage of target volumes (TVs) is also finding its way to RT centres worldwide.