Dear Humies Competition Committee, We would like to register for the 2019 Humies competition. The 11 items requested for our entry are detailed below: 1. The complete title of one (or more) paper(s) published in the open literature describing the work that the author claims describes a human-competitive result; Evaluation of bi-objective treatment planning for high-dose-rate prostate brachytherapy - A retrospective observer study 2. The name, complete physical mailing address, e-mail address, and phone number of EACH author of EACH paper(s); The following coauthors are affiliated with the Amsterdam UMC: Department of Radiation Oncology Amsterdam UMC, Location AMC University of Amsterdam Meibergdreef 9 1105 AZ Amsterdam The Netherlands S.C. Maree, MSc s.c.maree@amsterdamumc.nl Ernst S. Kooreman, MSc e.kooreman@nki.nl Niek van Wieringen, PhD n.vanwieringen@amsterdamumc.nl Arjan Bel, PhD a.bel@amsterdamumc.nl Karel A. Hinnen, MD, PhD k.a.hinnen@amsterdamumc.nl Henrike Westerveld, MD, PhD g.h.westerveld@amsterdamumc.nl Bradley R. Pieters, MD, PhD b.r.pieters@amsterdamumc.nl Tanja Alderliesten, PhD t.alderliesten@amsterdamumc.nl The following coauthors are affiliated with the national research institute for mathematics and computer science in the Netherlands (CWI): Life Science and Health Research Group Centrum Wiskunde & Informatica Science Park 123 1098 XG Amsterdam The Netherlands Ngoc Hoang Luong, PhD n.h.luong@cwi.nl Peter A.N. Bosman, PhD peter.bosman@cwi.nl 3. The name of the corresponding author (i.e., the author to whom notices will be sent concerning the competition); Stef Maree   +31205667973 s.c.maree@amsterdamumc.nl 4. The abstract of the paper(s); Purpose:  Bi-objective treatment planning for high-dose-rate prostate brachytherapy is a novel treatment planning method with two separate objectives that represent target coverage and organ-at-risk sparing. In this study, we investigated the feasibility and plan quality of this method by means of a retrospective observer study.  Methods and Materials: Current planning sessions were recorded to configure a bi-objective optimization model and to assess its applicability to our clinical practice. Optimization software, GOMEA, was then used to automatically generate a large set of plans with different trade-offs in the two objectives for each of 18 patients treated with high-dose-rate prostate brachytherapy. From this set, five plans per patient were selected for comparison to the clinical plan in terms of satisfaction of planning criteria and in a retrospective observer study. Three brachytherapists were asked to evaluate the blinded plans and select the preferred one.  Results:  Recordings demonstrated applicability of the bi-objective optimization model to our clinical practice. For 14/18 patients, GOMEA plans satisfied all planning criteria, compared with 4/18 clinical plans. In the observer study, in 53/54 cases, a GOMEA plan was preferred over the clinical plan. When asked for consensus among observers, this ratio was 17/18 patients. Observers highly appreciated the insight gained from comparing multiple plans with different trade-offs simultaneously.  Conclusions:  The bi-objective optimization model adapted well to our clinical practice. GOMEA plans were considered equal or superior to the clinical plans. In addition, presenting multiple high-quality plans provided novel insight into patient-specific trade-offs. 5. A list containing one or more of the eight letters (A, B, C, D, E, F, G, or H) that correspond to the criteria (see above) that the author claims that the work satisfies; D, G, H 6. A statement stating why the result satisfies the criteria that the contestant claims (see examples of statements of human-competitiveness as a guide to aid in constructing this part of the submission); (D) The result is publishable in its own right as a new scientific result independent of the fact that the result was mechanically created. The work was published both in computer science and in medical literature. Publication in the application domain indicates that the work has been accepted as a new scientific result, independent of our use of an evolutionary optimization approach to achieve it. Specifically, the clinical-evaluation study that we submit for this Humies competition was published in Brachytherapy, a journal which focuses on techniques and clinical applications of interstitial radiation (brachytherapy) in the management of cancer. Subsequent computational improvements to speed up the method were published in Radiotherapy & Oncology [1]. The technical details of the modelling of the problem were published in Swarm and Evolutionary Computation [2]. Furthermore, our bi-objective treatment planning approach is currently being implemented in the clinical routine in the Amsterdam UMC, location AMC. (G) The result solves a problem of indisputable difficulty in its field. Precise planning of the radiation dose is crucial for a successful application of high-dose-rate brachytherapy for prostate cancer. Multiple planning aims are formulated from clinical experience that result in a non-linear, non-smooth, and non-convex many-objective optimization problem. Until recently, it was considered computationally infeasible to perform optimization directly based on these planning aims in the clinical time limit. Current treatment planning methods optimize a simplified model of the planning problem, such that it can be quickly solved. However, the resulting solution is not always clinically acceptable. To overcome this, the parameters of the simplified model can be adapted, or the solution can be adapted directly by a technique called graphical optimization. Either approach makes the current clinical treatment planning process a largely-manual trial-and-error process that needs to be performed for each patient, requires experience, and is time-consuming, as Fig. 2 of our article nicely illustrates.  Instead of using a simplified model, we formulated a bi-objective optimization model [2] directly based on the planning aims combined into two objectives, resulting in a non-smooth, non-linear and non-convex bi-objective optimization problem. This model captures the most important trade-off of the problem: the trade-off between targeting the tumor and sparing the surrounding healthy tissue. We tailored the state-of-the-art multi-objective real-valued evolutionary algorithm GOMEA (Gene-pool Optimal Mixing Evolutionary Algorithm) to solve the bi-objective problem as efficiently as possible [1], and recent results showed that it takes only a few minutes to obtain a large set of high-quality solutions with different trade-offs in the two objectives. The final treatment plan can then be selected from this set, either automatically or manually. Furthermore, visualization of the obtained high-quality solutions as a patient-specific trade-off curve gives a unique and novel insight in the patient-specific trade-offs and can aid in the selection of the final plan. Using our approach, treatment planning can be transformed from a largely-manual trial-and-error process that takes over one hour, to an insightful decision making process that requires only a few minutes of computations [1]. (H) The result holds its own or wins a regulated competition involving human contestants (in the form of either live human players or human-written computer programs). Our article describes the clinical-evaluation study we performed, where we compared our bi-objective treatment planning approach with the current clinical practice. GOMEA was used to automatically generate treatment plans by optimizing the bi-objective model. Resulting plans were compared with clinical treatment plans that were used to actually treat the patient with, constructed with the largely-manual trial-and-error approach as described above. In a blinded experiment on 18 prostate cancer patients, per patient, three experienced physicians selected from five GOMEA plans and the clinical plan which plans were clinically acceptable, and which plan they preferred to treat the patient with. Physicians highly appreciated the patient-specific insight gained from being able to comparing multiple high-quality plans. Furthermore, this showed that our approach automatically generated clinically-acceptable treatment plans for all patients. In addition, for a stunning 98% of the cases, the automatically-generated plans were preferred over the clinical plan. 7. A full citation of the paper (that is, author names; publication date; name of journal, conference, technical report, thesis, book, or book chapter; name of editors, if applicable, of the journal or edited book; publisher name; publisher city; page numbers, if applicable); title = "Evaluation of bi-objective treatment planning for high-dose-rate prostate brachytherapy - A retrospective observer study", author = "S.C. Maree and N.H. Luong and E.S. Kooreman and N. van Wieringen and A. Bel and K.A. Hinnen and G.H. Westerveld and B.R. Pieters and P.A.N. Bosman and T. Alderliesten", journal = "Brachytherapy", volume = "18", number = "3", pages = "396 - 403", year = "2019", doi = "https://doi.org/10.1016/j.brachy.2018.12.010" 8. A statement either that "any prize money, if any, is to be divided equally among the co-authors" OR a specific percentage breakdown as to how the prize money, if any, is to be divided among the co-authors; Prize money, if any, is to be equally divided over the two collaborating research groups at CWI and Amsterdam UMC, represented by Tanja Alderliesten (Amsterdam UMC) and Peter Bosman (CWI). 9. A statement stating why the authors expect that their entry would be the "best" Prostate cancer is the second most common cancer in men, and the fifth leading cause of cancer death [3]. Among its treatment possibilities is high-dose-rate brachytherapy, a form of radiation therapy where the tumor is locally irradiated, which has great potential to cure the cancer while reducing side effects as much as possible. For a successful treatment, however, the to-be-delivered radiation dose needs to be distributed precisely. This process is called treatment planning. However, the current clinical treatment planning approach is time-consuming and requires experience, because current software methods are not able to automatically generate clinically-acceptable treatment plans, and a manual trial-and-error process is required to overcome this. Our bi-objective treatment planning approach transforms treatment planning from this largely-manual trial-and-error approach into an insightful decision-making process. Driven by the state-of-the-art evolutionary algorithm GOMEA, specifically tuned to the treatment planning problem, we reduced plan construction time by over 90%, from largely difficult manual trial-and-error parameter tuning, to less than a few minutes of just computation. The bi-objective problem formulation makes the resulting decision-making process insightful. Experienced physicians considered plans generated by our method as clinically acceptable for all patients, and in a stunning 98% of the cases, they preferred plans generated by our method over plans constructed with the largely-manual current clinical treatment planning approach. 10. An indication of the general type of genetic or evolutionary computation used, such as GA (genetic algorithms), GP (genetic programming), ES (evolution ), EP (evolutionary programming), LCS (learning classifier systems), GE (grammatical evolution), GEP (gene expression programming), DE (differential evolution), etc. EDA - Estimation of Distribution Algorithm 11. The date of publication of each paper. If the date of publication is not on or before the deadline for submission, but instead, the paper has been unconditionally accepted for publication and is in "press" by the deadline for this competition, the entry must include a copy of the documentation establishing that the paper meets the "in press" requirement. Available online 2 February 2019. References [1] A. Bouter, T. Alderliesten, B.R. Pieters, A. Bel, Y. Niatsetski, P.A.N. Bosman, OC-0395 Bi-objective optimization of dosimetric indices for HDR prostate brachytherapy within 30 seconds. Radiotherapy and Oncology 133, S199-S200 (2019) [2] N.H. Luong, T. Alderliesten, A. Bel, Y. Niatsetski, P.A.N. Bosman. Application and benchmarking of multi-objective evolutionary algorithms on high-dose-rate brachytherapy planning for prostate cancer treatment. Swarm and Evolutionary Computation 40, 37-52 (2018) [3] International Agency for research on Cancer, Cancer Fact Sheets: prostate cancer, 2016, gco.iarc.fr/today/data/pdf/fact-sheets/cancers/cancer-fact-sheets-19.pdf