<div dir="ltr"><h1 style="margin-right:12pt;margin-left:0in;font-size:24pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);margin-bottom:3pt;vertical-align:baseline"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51);font-weight:normal">Focus on Machine Learning Models in Medical Imaging</span></h1><h3 style="margin-right:0in;margin-left:0in;font-size:2rem;font-family:Calibri,sans-serif;color:rgb(0,0,0);font-style:inherit;font-variant:inherit;margin-bottom:3pt;vertical-align:baseline;font-stretch:inherit"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Guest Editors</span></h3><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin:0in;vertical-align:baseline"><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Dr Giorgos Papanastasiou</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">, University of Essex, UK<br></span><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Dr Alba García Seco de Herrera</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">, University of Essex, UK<br></span><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Dr Chengjia Wang</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">, University of Edinburgh, UK<br></span><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Prof Heye Zhang</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">, Sun Yat-sen University, China<br></span><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Dr Guang Yang</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">, Imperial College London, UK<br></span><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Prof Ge Wang</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">, Rensselaer Polytechnic Institute, USA</span></p><h3 style="margin-right:0in;margin-left:0in;font-size:2rem;font-family:Calibri,sans-serif;color:rgb(0,0,0);font-style:inherit;font-variant:inherit;margin-bottom:3pt;vertical-align:baseline;font-stretch:inherit"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Scope</span></h3><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin-right:0in;margin-bottom:12pt;margin-left:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Computational medical imaging techniques aim to enhance the diagnostic performance of visual assessments in medical imaging, improving the early diagnosis of various diseases and helping to obtain a deeper understanding of physiology and pathology.</span></p><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin-right:0in;margin-bottom:12pt;margin-left:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Machine Learning (ML) models revolutionised multiple tasks in medical image computing, such as image segmentation, registration and synthesis, through the extensive analysis of big imaging data. Although ML models outperform classic approaches on these tasks, they remain to a large extend implicit in terms of describing the data under investigation. This limits ML model interpretability, which is one of the main barriers towards ML-based pathology detection and generalised single- or multi-modal ML analysis in medical imaging. In modern clinical practices, detailed explanations of the model behaviours are increasingly required to support reliability towards improving clinical decision making. Moreover, being of the most promising topics in ML/medical imaging research, the main challenge for developing explainable models is to offer insights and rationales whilst maintaining high learning performance.</span></p><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">With this joint focus issue, between <a href="https://iopscience.iop.org/journal/0031-9155" style="color:rgb(5,99,193)"><em><span style="color:rgb(0,110,178);border:1pt none windowtext;padding:0in">Physics in Medicine and Biology</span></em></a> and the multidisciplinary open access journal <a href="https://iopscience.iop.org/journal/2632-2153" style="color:rgb(5,99,193)"><em><span style="color:rgb(0,110,178);border:1pt none windowtext;padding:0in">Machine Learning: Science and Technology</span></em></a>, we aim to attract original high-quality research and survey articles that reflect the most recent advances on ML models in medical imaging (MRI, CT, PET, SPECT, Ultrasound and other), by investigating novel methodologies either through interpreting algorithm components and/or exploring algorithm-data relationships.</span></p><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin-right:0in;margin-bottom:12pt;margin-left:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Articles will be published in one of two participating journals and we are happy to let authors choose which journal to submit to based on the criteria below. If an article submitted to one journal is found to be unsuitable for consideration, but suitable for the other, the assessing Editor will offer the author an opportunity to transfer their article. This means that duplication of peer review effort can be largely eliminated as a service to our authors.</span></p><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin:0in;vertical-align:baseline"><em><b><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Physics in Medicine and Biology</span></b></em><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)"> encourages the submission of papers that focus on the medical interpretation, clinical impact, applications and modalities and </span><em><b><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">Machine Learning: Science and Technology</span></b></em><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)"> encourages the submission of papers that focus on the methodology and physics-based interpretation of the technical aspects of machine learning models.</span></p><h3 style="margin-right:0in;margin-left:0in;font-size:2rem;font-family:Calibri,sans-serif;color:rgb(0,0,0);font-style:inherit;font-variant:inherit;margin-bottom:3pt;vertical-align:baseline;font-stretch:inherit"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Topics:</span></h3><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin-right:0in;margin-bottom:12pt;margin-left:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">We welcome researchers from academia, clinics and industry, to present their state-of-the-art scientific developments covering all aspects of ML model in medical imaging.</span></p><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin-right:0in;margin-bottom:12pt;margin-left:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Potential topics include but are not limited to:</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Develop and interpret ML models in single- or multi-modal (MRI, CT, Ultrasound, PET, SPECT) imaging</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Multi-task learning on multi-modality medical images</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Solidify explainability in cross-domain image synthesis between different imaging modalities or sequences (e.g. from different MRI sequences, or MRI and CT, etc.)</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Transfer learning for single- or multi-modality medical images</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">ML model explainability in semi-supervised, weakly-supervised and unsupervised learning in medical imaging</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Enhance explainability through developing ML models to detect or predict pathology versus healthy statuses</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">To improve explainability, combine ML with biophysical modelling and/or visual assessments from additional/complementary imaging modalities (e.g., multiple sequences in MRI, or combining MRI with Ultrasound, CT, PET or SPECT)</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">To improve explainability, combine ML with other types of "reference standard" input data (e.g. clinical data, electrophysiology signals, molecular analysis, invasive methods) that can enhance ML interpretability in medical imaging</span></p><p class="MsoNormal" style="margin:0in 0in 0in 57pt;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);vertical-align:baseline"><span style="font-size:10pt;color:rgb(51,51,51)">·<span style="font-stretch:normal;font-size:7pt;line-height:normal;font-family:"Times New Roman"">        <span class="gmail-Apple-converted-space"> </span></span></span><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Explain strengths and weaknesses of ML models through quantitative evaluation and interpretation of algorithm performance, especially mechanisms of adversarial attacks and associated solutions</span></p><h3 style="margin-right:0in;margin-left:0in;font-size:2rem;font-family:Calibri,sans-serif;color:rgb(0,0,0);font-style:inherit;font-variant:inherit;margin-bottom:3pt;vertical-align:baseline;font-stretch:inherit"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Deadline for submissions</span></h3><p style="color:rgb(0,0,0);font-family:-webkit-standard;margin:0in;vertical-align:baseline"><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Submissions will be accepted until </span><strong><span style="font-size:12pt;color:rgb(51,51,51);border:1pt none windowtext;padding:0in">31 August 2021</span></strong><span style="font-size:12pt;font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)"> however submissions earlier than this date are encouraged.</span></p><p class="MsoNormal" style="margin:0in;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0)"><span style="font-size:11pt"> </span></p><h3 style="margin-right:0in;margin-left:0in;font-size:13.5pt;font-family:Calibri,sans-serif;color:rgb(0,0,0);margin-bottom:3pt;vertical-align:baseline"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">More information</span></h3><p class="MsoNormal" style="margin:0in;font-size:12pt;font-family:Calibri,sans-serif;color:rgb(0,0,0)"><span style="font-family:"Segoe UI",sans-serif;color:rgb(51,51,51)">Submissions will be<span class="gmail-Apple-converted-space"> </span><a href="https://iopscience.iop.org/journal/0031-9155/page/Focus%20on%20Machine%20Learning%20Models%20in%20Medical%20Imaging" title="https://iopscience.iop.org/journal/0031-9155/page/Focus%20on%20Machine%20Learning%20Models%20in%20Medical%20Imaging" style="color:rgb(5,99,193)">https://iopscience.iop.org/journal/0031-9155/page/Focus%20on%20Machine%20Learning%20Models%20in%20Medical%20Imaging</a></span></p></div>