{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T15:00:36Z","timestamp":1779894036794,"version":"3.53.1"},"reference-count":46,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2023,9,8]],"date-time":"2023-09-08T00:00:00Z","timestamp":1694131200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2024,3]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Real\u2010time tumor tracking is one motion management method to address motion\u2010induced uncertainty. To date, fiducial markers are often required to reliably track lung tumors with X\u2010ray imaging, which carries risks of complications and leads to prolonged treatment time. A markerless tracking approach is thus desirable. Deep learning\u2010based approaches have shown promise for markerless tracking, but systematic evaluation and procedures to investigate applicability in individual cases are missing. Moreover, few efforts have been made to provide bounding box prediction and mask segmentation simultaneously, which could allow either rigid or deformable multi\u2010leaf collimator\u00a0tracking.<\/jats:p><\/jats:sec><jats:sec><jats:title>Purpose<\/jats:title><jats:p>The purpose of this study was to implement a deep learning\u2010based markerless lung tumor tracking model exploiting patient\u2010specific training which outputs both a bounding box and a mask segmentation simultaneously. We also aimed to compare the two kinds of predictions and to implement a specific procedure to understand the feasibility of markerless tracking on individual\u00a0cases.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We first trained a Retina U\u2010Net baseline model on digitally reconstructed radiographs (DRRs) generated from a public dataset containing 875 CT scans and corresponding lung nodule annotations. Afterwards, we used an independent cohort of 97 lung patients to develop a patient\u2010specific refinement procedure. In order to determine the optimal hyperparameters for automatic patient\u2010specific training, we selected 13 patients for validation where the baseline model predicted a bounding box on planning CT (PCT)\u2010DRR with intersection over union (IoU) with the ground\u2010truth higher than 0.7. The final test set contained the remaining 84 patients with varying PCT\u2010DRR IoU. For each testing patient, the baseline model was refined on the PCT\u2010DRR to generate a patient\u2010specific model, which was then tested on a separate 10\u2010phase 4DCT\u2010DRR to mimic the intrafraction motion during treatment. A template matching algorithm served as benchmark model. The testing results were evaluated by four metrics: the center of mass (COM) error and the Dice similarity coefficient (DSC) for segmentation masks, and the center of box (COB) error and the DSC for bounding box detections. Performance was compared to the benchmark model including statistical testing for\u00a0significance.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>A PCT\u2010DRR IoU value of 0.2 was shown to be the threshold dividing inconsistent (68%) and consistent (100%) success (defined as mean bounding box DSC &gt; 0.6) of PS models on 4DCT\u2010DRRs. Thirty\u2010seven out of the eighty\u2010four testing cases had a PCT\u2010DRR IoU above 0.2. For these 37 cases, the mean COM error was 2.6 mm, the mean segmentation DSC was 0.78, the mean COB error was 2.7 mm, and the mean box DSC was 0.83. Including the validation cases, the model was applicable to 50 out of 97 patients when using the PCT\u2010DRR IoU threshold of 0.2. The inference time per frame was 170 ms. The model outperformed the benchmark model on all metrics, and the comparison was significant (<jats:italic>p<\/jats:italic> &lt; 0.001) over the 37 PCT\u2010DRR IoU &gt; 0.2 cases, but not over the undifferentiated 84 testing\u00a0cases.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>The implemented patient\u2010specific refinement approach based on a pre\u2010trained baseline model was shown to be applicable to markerless tumor tracking in simulated radiographs for lung\u00a0cases.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.16705","type":"journal-article","created":{"date-parts":[[2023,9,8]],"date-time":"2023-09-08T19:20:32Z","timestamp":1694200832000},"page":"1957-1973","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":10,"title":["Simultaneous object detection and segmentation for patient\u2010specific markerless lung tumor tracking in simulated radiographs with deep learning"],"prefix":"10.1002","volume":"51","author":[{"given":"Lili","family":"Huang","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"},{"name":"Department of Medical Physics Faculty of Physics Ludwig\u2010Maximilians\u2010Universit\u00e4t M\u00fcnchen M\u00fcnchen Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Kurz","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Philipp","family":"Freislederer","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Farkhad","family":"Manapov","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stefanie","family":"Corradini","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maximilian","family":"Niyazi","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Claus","family":"Belka","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"},{"name":"German Cancer Consortium (DKTK), partner site Munich, a partnership between DKFZ and LMU University Hospital Munich Germany"},{"name":"Bavarian Cancer Research Center (BZKF) Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guillaume","family":"Landry","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology LMU University Hospital LMU Munich Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marco","family":"Riboldi","sequence":"additional","affiliation":[{"name":"Department of Medical Physics Faculty of Physics Ludwig\u2010Maximilians\u2010Universit\u00e4t M\u00fcnchen M\u00fcnchen Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2023,9,8]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.1531177"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/48\/1\/304"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/0360-3016(89)90078-3"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.1287110"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0360-3016(00)00681-7"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0360-3016(00)00583-6"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.3109\/02841869409121782"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12040"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/46\/1\/301"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/58\/12\/4195"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aa6393"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4918578"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12073"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0360-3016(00)00748-3"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(99)00700-X"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.gie.2012.07.006"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2013.05.001"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00270-010-9949-0"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4917480"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aa6393"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2013.06.2048"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2019.05.071"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2019.06.027"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20190420"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab79c5"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15456"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab6f51"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12731"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14445"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.1088\/1742-6596\/489\/1\/012079"},{"key":"e_1_2_9_32_1","unstructured":"JaegerPF KohlSAA BickelhauptS et\u00a0al.Retina U\u2010Net: embarrassingly simple exploitation of segmentation supervision for medical object detection.Proceedings of the Machine Learning for Health NeurIPS Workshop.PMLR;2020:171\u2010183."},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-94750-z"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.3632567"},{"key":"e_1_2_9_35_1","doi-asserted-by":"crossref","unstructured":"SimardP SteinkrausD PlattJ.Best practices for convolutional neural networks applied to visual document analysis.Seventh International Conference on Document Analysis and Recognition 2003. Proceedings. Edinburgh UK 2003:958\u2010963.https:\/\/doi.org\/10.1109\/ICDAR.2003.1227801","DOI":"10.1109\/ICDAR.2003.1227801"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.453"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab2667"},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14536"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4802735"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac60b7"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.3390\/app8122525"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2021.100705"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1088\/0957-0233\/24\/7\/074012"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aada71"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/60\/24\/9437"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.20965\/jaciii.2022.p0471"},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15418"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.16705","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,12]],"date-time":"2024-03-12T15:06:18Z","timestamp":1710255978000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.16705"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,8]]},"references-count":46,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2024,3]]}},"alternative-id":["10.1002\/mp.16705"],"URL":"https:\/\/doi.org\/10.1002\/mp.16705","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,8]]},"assertion":[{"value":"2022-09-27","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-05-12","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-09-08","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}