<div><div><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><b style="font-family:"times new roman",serif"><span style="font-size:14pt;line-height:1.5;font-family:"times new roman",serif;color:black">Submission deadline <span dir="ltr" style="text-decoration:underline;font-family:"times new roman",serif;font-size:inherit!important">24</span><sup style="font-family:"times new roman",serif"><span dir="ltr" style="text-decoration:underline;font-family:"times new roman",serif;font-size:inherit!important">th</span></sup><span dir="ltr" style="text-decoration:underline;font-family:"times new roman",serif;font-size:inherit!important"> June 2024</span> - </span></b><span style="font-size:14pt;line-height:1.5;font-family:"times new roman",serif"><a href="https://unsuremiccai.github.io/" style="text-decoration:underline;font-family:"times new roman",serif;color:black" target="_blank"><b style="font-family:"times new roman",serif"><span style="font-family:"times new roman",serif;color:rgb(17,85,204)">https://unsuremiccai.github.io</span></b></a></span></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><b style="font-family:"times new roman",serif"><u style="font-family:"times new roman",serif"><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:rgb(17,85,204)"><br></span></u></b><b style="font-family:"times new roman",serif"><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:black">Overview<br><br></span></b><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:black">With the rise and influence of machine learning (ML) in medical application and the need to translate newly developed techniques into clinical practice, questions about safety and uncertainty over measurements and reported quantities have gained importance. Obtaining accurate measurements is insufficient, as one needs to establish the circumstances under which these values generalize, or give appropriate error bounds for these measures. This is becoming particularly relevant to patient safety as many research groups and companies have deployed or are aiming to deploy ML technology in clinical practice.<br><br>The purpose of this workshop is to develop awareness and encourage research on uncertainty modelling to ensure safety for applications spanning both the MIC and CAI fields. In particular, this workshop invites submissions to cover different facets of this topic, including but not limited to: detection and quantification of algorithmic failures; processes of healthcare risk management (e.g. CAD systems); robustness and adaptation to domain shifts; evaluation of uncertainty estimates; defence against noise and mistakes in data (e.g. bias, label mistakes, measurement noise, inter/intra-observer variability). The workshop aims to encourage contributions in a wide range of applications and types of ML algorithms. The use or development of any relevant ML methods are welcomed, including, but not limited to, probabilistic deep learning, Bayesian nonparametric statistics, graphical models and Gaussian processes. We also aim to ensure broad coverage of applications in the context of both MIC and CAI, which are categorized into reporting problems (descriptions of image contents) such as diagnosis, measurements, segmentation, detection, and enhancement problems (addition of information) such as image synthesis, registration, reconstruction, super-resolution, harmonisation, inpainting and augmented display.</span></p><p style="margin-right:0in;margin-left:0in;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;margin-bottom:12pt;color:rgb(0,0,0)"><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:black"> </span></p><p style="margin-right:0in;margin-left:0in;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;margin-bottom:15pt;color:rgb(0,0,0)"><b style="font-family:"times new roman",serif"><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:black">Scope</span></b></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-family:"times new roman",serif;color:black">We accept submissions of original, unpublished work on <b style="font-family:"times new roman",serif">safety and uncertainty in medical imaging</b>, including (but not limited to) the following areas:</span></p><ul type="disc" style="margin-bottom:0in;font-family:-apple-system,helveticaneue;font-size:16px;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;margin-top:0in;color:rgb(0,0,0)"><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Uncertainty quantification in any MIC or CAI applications </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Risk management of ML systems in clinical pipelines </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Out-of-distribution and anomaly detection </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Defending against hallucinations in enhancement tasks (e.g. super-resolution, reconstruction, modality translation) </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Robustness to domain shifts </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Measurement errors </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Modelling noise in data (e.g. labels, measurements, inter/intra-observer variability) </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Validation of uncertainty estimates </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Active Learning </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Confidence bounds </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Posterior inference over point estimates </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Bayesian deep learning</span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Graphical models </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Gaussian processes </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Calibration of uncertainty measures </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li><li style="margin:0in 0in 0.0001pt;font-size:12pt;font-family:aptos;vertical-align:baseline;color:black"><span style="font-size:11pt;font-family:"times new roman",serif;line-height:1.5">Bayesian decision theory </span><span style="font-size:11pt;font-family:arial,sans-serif;line-height:1.5"></span></li></ul><p style="margin-right:0in;margin-left:0in;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;margin-bottom:15pt;color:rgb(0,0,0)"><b style="font-family:"times new roman",serif"><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:black"> </span></b></p><p style="margin-right:0in;margin-left:0in;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;margin-bottom:15pt;color:rgb(0,0,0)"><b style="font-family:"times new roman",serif"><span style="font-size:11pt;line-height:1.5;font-family:"times new roman",serif;color:black">Submission Format</span></b></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-family:"times new roman",serif;color:black">Submissions must be 8-page papers <b style="font-family:"times new roman",serif">(excluding references)</b> following the </span><a href="https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines" style="text-decoration:underline;font-family:"times new roman",serif;color:black" target="_blank"><span style="font-family:"times new roman",serif;color:rgb(17,85,204)">Springer LNCS format</span></a><span style="font-family:"times new roman",serif;color:black">. Author names, affiliations and acknowledgements, as well as any obvious phrasings or clues that can identify authors must be removed to ensure anonymity. Note that the 8 page limit refers only to the main content. Including references and acknowledgements the submission may exceed 8 pages.</span></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-family:"times new roman",serif;color:black"> </span></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-family:"times new roman",serif;color:black">Please submit papers using the </span><a href="https://cmt3.research.microsoft.com/UNSURE2024/Submission/" style="text-decoration:underline;font-family:"times new roman",serif;color:black" target="_blank"><span style="font-family:"times new roman",serif;color:rgb(17,85,204)">paper submission system</span></a></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-family:arial,sans-serif;color:black"> </span></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:"times new roman",serif;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-family:"times new roman",serif;color:black">We plan to publish the proceedings as an LNCS volume. Accepted papers will also be invited for submission of an extended version to the MELBA journal as part of a special issue. </span></p><p style="margin:0in 0in 0.0001pt;font-size:16px;font-family:aptos;font-style:normal;font-weight:400;letter-spacing:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;text-decoration:none;color:rgb(0,0,0)"><span style="font-size:11pt;line-height:1.5;font-family:aptos"> </span></p></div>
</div>