\section{Method}

We conduct our work according to the methodology used by~\cite{Mentis2014}. The examples presented in this paper
come from the
Champalimaud Foundation in Lisbon, PT~\cite{Zorzal2020}.
At that site, we
observed laparoscopic surgeries on one of the Foundation's surgical rooms and
talked to surgeons before and after the procedures, who explained
what was about to happen or what took place. Also, during the surgery, nurses
provided insight into the several stages of the surgery, or what was happening at that time. We also had the opportunity to ask questions of the surgeons at appropriate moments during the surgery. Besides that, we video recorded for further analysis in addition to our field notes. 
A total of five (5) laparoscopic surgeries were observed for a total of approximately 10 hours.

The observed cases included the surgical team is composed of mostly male members, with only one female surgeon in it, and their ages range from 34 to 42 years of age.
During surgery, there are at least six people involved in the procedure
A head surgeon, who coordinates the entire procedure, 1 or 2 assistant surgeons, who mostly observe but also participate in parts of the surgery, a nurse solely responsible for passing the surgeons tools they may require throughout the operation, an anesthetist keeping track of the patient's vital signs, a nurse supporting the anesthetist and a circulating nurse. Additionally, a senior surgeon may come in and serve as an advisor, providing insight and making remarks about what is being seen on camera.


\bgroup




We collected images and videos using smartphones and tablets, as well as the notes made during the observations. The inductive bottom-up approach to data analysis was used in which the authors analyzed their field notes and videos. The findings and the discussion are presented in the subsequent section. 

\section{Results and Discussion}

















Performing user and task analysis allowed us to better understand the existing problems in the procedure of laparoscopy, while identifying several constraints and design requirements, which a solution has to follow in order to address those problems. For the following, we will discuss the design requirements identified through the analysis. Problems statements, requirements, and the proposed design solutions also are summarized in Table~\ref{table:problem_requirements_solution}.




\begin{table*}[htb]
\caption{Problem statements, design requirements and design solution for a prototype that supports laparoscopy}\label{table:problem_requirements_solution}
\begin{tabular}{|p{0.31\textwidth}|p{0.31\textwidth}|p{0.31\textwidth}|}
\hline
{\bfseries \small Problem Statements} & {\bfseries \small Design Requirements} & {\bfseries \small Design Solution}\\
\hline
\small Visualizing the laparoscopic video during extended periods of time is exhausting for the neck. & \small The solution should allow the user to adopt more comfortable neck postures instead of forcing the user to look to the side to see what the other surgeons are seeing. & \small Following display: The laparoscopic video follows user head movement, so users can look around and assume a neck posture that is more comfortable for them.  \\
\hline
    \small Current interactions surgeons have, such as pointing or consulting patient data, require them to let go of their tools, which interrupts the procedure. & \small Surgeons should have hands-free interactions in order to operate in an uninterrupted fashion. & \small Hands-free interaction: Every interaction is either done with the head or using the feet. \\
    \hline
    \small Browsing patient data interoperatively takes too long because it requires to call in an assistant, who browses the images for the surgeon. & \small Users should be able to look at patient data by themselves, without interrupting and adding extra time to the surgery. & \small Patient data image browser: users can look to the side to see and browse magnetic resonance images from the patient.  \\
    \hline
    \small Users may have to move around the patient in order to adopt better positions to hold their tools. & \small Interaction using the foot should not rely on pedals, as these would need to be moved around to cope with user movement. & \small Foot browsing: Users can use the foot to navigate the patient images, rotating it on its heel to change images faster or slower. \\
    \hline
    \small Pointing is unclear and ambiguous: different users have different interpretations of where a surgeon is pointing at. & \small Users should be able to point precisely and understand where other users are pointing at, regardless of position in the operating room. & \small Pointing reticle: users can place a reticle on both laparoscopic video and patient images, controlling it with head motion. This cursor is visible on other users’ headsets.\\
    \hline
    \small Surgeons operate in a crowded area, as they are usually very close together. & \small Augmented space should present information close to the surgeon to prevent it from appearing intersected with a colleague. & \small Close quarters: Positioning of interface elements is no further than at an elbow's reach.\\
    \hline
\end{tabular}
\end{table*}



\subsection{Following display}

Laparoscopy is an intensive process, not just mentally but physically as well. The procedure is already very demanding in itself due to surgeons having to expend extra mental effort thanks to a lack of hand-eye coordination that is caused by indirect visualization (Fig.~\ref{fig:neck}). That effort extends to the physical plane when we consider that they have look at the screen all the time, which places a continuous strain on their necks. Laparoscopy currently faces the glaring problem of monitor positioning.  During surgery, screens are usually placed far away and at an uncomfortable angle, causing neck and eye strain over the course of a surgery, especially if it drags for longer periods of time. Given this, it was important to allow the surgeons some freedom in how they want to see the video, and then the video,  while visible,  should follow user head movements so users do not have to reposition it in the augmented space, should they feel the need to assume another posture with the neck. 




\begin{figure*}[!htbp]
    \centering
    \includegraphics[width=0.65\textwidth,keepaspectratio]{Figure_3.png}
    \caption{All doctors look to the same screen and, sometimes, this means having to assume an uncomfortable position. Visualizing the laparoscopic video during extended periods of time is exhausting for the neck.}
    \label{fig:neck}
\end{figure*} 

The surgeons have to look at screens placed outside the field of operation, which results in discomfort~\cite{Batmaz2017}, affecting the surgeon's efficiency due to a disconnect between the visual and motor axis, because the surgeon cannot look at the instruments or hands and the field of surgery simultaneously. To be successful, more training is required to adapt to this condition, as extra mental effort must be applied~\cite{Leite2016}.
In addition, almost all
display screens are limited in sense that they do not support techniques to improve communication and visual collaboration with the rest of the surgical team~\cite{HenryFuchs1MarkA.Livingston1RameshRaskar1DnardoColucci11963,HMentis2019}.

Muratore et al.~\cite{Muratore2007}  suggest for the future,
the ideal display system would be a 3D high definition image HMD,
citing the comfort of looking at the endoscopic image in any preferred head position, improving ergonomics and reducing neck strain. The use of an HMD is also seen as beneficial in the sense that it alleviates equipment clutter in the operating room. It is further noted the usefulness of individualized image manipulation features like zooming, which allows each surgeon to see the endoscopic video in the way they find most comfortable.
Also, the works of~\cite{Walczak2015,Maithel2005,Batmaz2017,Kihara2012} seem to support the usage of a HMD for laparoscopic surgery, with the video following the user’s head movements. With this, users can assume their preferred head position instead of being forced to look sideways in order to see the video. 


\subsection{Hands-free interaction}

From our field observations, we noticed that surgeons place down their tools to perform some secondary tasks, which interrupts the procedure (Fig.~\ref{fig:losing}). We suggest that projects avoid this type of situation, with a completely hands-free approach, using both head gaze and foot movement as sources of input.




\begin{figure*}[!bhtpb]
    \centering
    \includegraphics[width=1\textwidth,keepaspectratio]{Figure_4.png}
    \caption{Surgeons can't look at their hands, thus losing hand-eye coordination.}
    \label{fig:losing}
\end{figure*}

Other studies~\cite{Kim2017,Jayender}
looked at
using head movements, and gaze
to select targets, especially
when the content follows head movements~\cite{Grinshpoon}. A more elaborate approach takes the form of the eye gazing~\cite{Esteves2015,Velloso2016}, which was well-received by users, but may not transition well onto the surgical operating field: these controls would have to be displayed continuously and right in front of the user, unlike in the presented works, which could be distractive for users, but more importantly, it would take valuable space from the HMD's already limited field of view. In terms of feet, two different approaches emerge: using a foot pedal as a means to activate a selected control~\cite{Jayender} and using foot movement to select and activate controls~\cite{Muller2019}. After comparing the two, we conclude using foot movement would be a more flexible choice, as it does not rely on extra hardware that is situated in a given position in space.

\subsection{Patient data image browser and foot browsing}


Consulting patient data intra-operatively, such as MRIs and computed tomographies, may be unfeasible. On the one hand, the data are extensive and may require some time to identify the required set. On the other, surgeons must abandon the operating table to sit at a computer and browse the desired images. The surgeon then gives directions on where to look and when to stop while an assistant handles the computer.  Surgeons usually do not browse the images themselves when sterilized. Indeed each interaction with non-sterile equipment such as the keyboard and mouse would entail a new sterilization procedure, wasting additional resources. This has been discussed in \cite{ohara-2014,Lopes-2019,Johnson2011}. This is a design opportunity for AR as discussed in \cite{Zorzal2019}.

In the work of
\cite{Muratore2007}, the authors emphasize the importance of the HMD, stating preoperative imaging could be individually manipulated through its use, as well as grant surgeons extra comfort by allowing them to see the laparoscopic image regardless of head positioning. Hands-free interaction is again considered, with the suggestion of using foot pedals instead. Also discussed is the issue of paradoxical imaging, which occurs when the camera faces the surgeon, causing movements with the tool to appear inverted compared to the hand movements. However, in an interview with the surgeons of the
Champalimaud Foundation,  this did not appear to be an issue, since the surgeon can move around the
operating table, which ensures the camera always faces the opposite direction. This freedom to move around also impacts the practicality of using foot pedals to ensure hands-free interaction, as the authors suggest, as the surgeon would have to either have the same pedals on multiple sides, or move the pedals around. In this case, exploring foot movement, as proposed by
\cite{Muller2019}, could be more useful




\subsection{Pointing reticle and close quarters}

Although they are close quarters, surgeons also currently face problems in communication. In fact, according to the inquired surgeons, just as they complain about difficulty in maintaining proper posture, so do they complain about not being able to let other surgeons know what part of the video they are pointing at, or to understand what others are pointing at as well. Several works have approached this by looking at proxemics \cite{Mentis2012} and embodied vision~\cite{Mentis2013}. 




   

\begin{figure*}[!htpb]

    \centering
    \includegraphics[width=0.65\textwidth,keepaspectratio]{Figure_6.png}
    \caption{Doctor point at the screen to communicate. Communication is unclear and ambiguous: different users have different interpretations of where a surgeon is pointing at.}
    \label{fig:pointing2}
\end{figure*}




The instructor can point at the screen 
for the other surgeons to understand what anatomical structure he/she is referring to and
use gestures for others to understand the motion of the tools better and envision cutting lines. Sometimes, pointing can also be done with the tools themselves, but even though it may be effective, it is not always correct because if both hands are occupied, it implies letting go of a structure to point with the tool or asking someone else to hold it. Additionally, pointing from a distance with the hand is ambiguous at best, as there is no clear way to tell where precisely a surgeon is pointing at, as can be seen in Fig.~\ref{fig:pointing2}.


For Prescher et al.~\cite{Prescher}, the impact of the pointer in a real operating scenario may be lessened because target selection is not random but rather contextual, meaning that the following targets may be located through the description of what is being displayed on-screen. Also, the pointer is embedded in the laparoscope. Thus the camera must be displaced to move the pointer, causing the view plane to change and forcing the surgeon to readjust to the new perspective, losing perceived depth. Therefore, it would be more useful if the cursor moved independently of the camera, controlled via gestures or head tracking for a hands-free approach, as suggested by 3D interactions performed above a table~\cite{mendes2014,mendes2016}.






\section{Conclusion}



This paper presents the requirements and design solutions for laparoscopy through a user and task analysis. During a task analysis, we map out the sequence of activities surgeons go through and the actions required to achieve that goal. Drawing on observations and analysis of video recordings of laparoscopic surgeries, we identify several constraints and design requirements, which a solution will have to follow in order to address those problems. These requirements propose to inform the design solutions towards improved surgeons' comfort and make the surgical procedure less laborious.



\begin{comment}
\section*{Summary points} 
\noindent
\setlength{\fboxsep}{10pt}
\fbox{\parbox{0.9\columnwidth}{\textbf{What is already known on the topic?}
\begin{itemize}
\setlength\itemsep{0pt}
\item Laparoscopy suffers some problems that can cause the surgical procedure more laborious.
\item A variety of different problems influence the performance of surgeons in laparoscopic procedures.
\end{itemize}
\textbf{What does this study add to our knowledge?}
\begin{itemize}
\setlength\itemsep{0pt}
\item Identification of problems related to design for laparoscopy.
\item Design requirements propose to inform the design solutions towards improved surgeons' comfort and make the surgical procedure less laborious.


\end{itemize}

}}

\end{comment}



\section*{Acknowledgments}





This work was supported by national funds through FCT, Fundação para a Ciência e a Tecnologia, under project UIDB/50021/2020. The authors would like to thank the Champalimaud Foundation for its collaboration and support in the development of the study.








\bibliographystyle{abbrv-doi}


\section{Introduction}
This template is for papers of VGTC-sponsored conferences which are \emph{\textbf{not}} published in a special issue of TVCG.

\section{Using the Style Template}

\begin{itemize}
\item If you receive compilation errors along the lines of ``\texttt{Package ifpdf Error: Name clash, \textbackslash ifpdf is already defined}'' then please add a new line ``\texttt{\textbackslash let\textbackslash ifpdf\textbackslash relax}'' right after the ``\texttt{\textbackslash documentclass[journal]\{vgtc\}}'' call. Note that your error is due to packages you use that define ``\texttt{\textbackslash ifpdf}'' which is obsolete (the result is that \texttt{\textbackslash ifpdf} is defined twice); these packages should be changed to use ifpdf package instead.
\item The style uses the hyperref package, thus turns references into internal links. We thus recommend to make use of the ``\texttt{\textbackslash autoref\{reference\}}'' call (instead of ``\texttt{Figure\~{}\textbackslash ref\{reference\}}'' or similar) since ``\texttt{\textbackslash autoref\{reference\}}'' turns the entire reference into an internal link, not just the number. Examples: \autoref{fig:sample} and \autoref{tab:vis_papers}.
\item The style automatically looks for image files with the correct extension (eps for regular \LaTeX; pdf, png, and jpg for pdf\LaTeX), in a set of given subfolders (figures/, pictures/, images/). It is thus sufficient to use ``\texttt{\textbackslash includegraphics\{CypressView\}}'' (instead of ``\texttt{\textbackslash includegraphics\{pictures/CypressView.jpg\}}'').
\item For adding hyperlinks and DOIs to the list of references, you can use ``\texttt{\textbackslash bibliographystyle\{abbrv-doi-hyperref-narrow\}}'' (instead of ``\texttt{\textbackslash bibliographystyle\{abbrv\}}''). It uses the doi and url fields in a bib\TeX\ entry and turns the entire reference into a link, giving priority to the doi. The doi can be entered with or without the ``\texttt{http://dx.doi.org/}'' url part. See the examples in the bib\TeX\ file and the bibliography at the end of this template.\\[1em]
\textbf{Note 1:} occasionally (for some \LaTeX\ distributions) this hyper-linked bib\TeX\ style may lead to \textbf{compilation errors} (``\texttt{pdfendlink ended up in different nesting level ...}'') if a reference entry is broken across two pages (due to a bug in hyperref). In this case make sure you have the latest version of the hyperref package (i.\,e., update your \LaTeX\ installation/packages) or, alternatively, revert back to ``\texttt{\textbackslash bibliographystyle\{abbrv-doi-narrow\}}'' (at the expense of removing hyperlinks from the bibliography) and try ``\texttt{\textbackslash bibliographystyle\{abbrv-doi-hyperref-narrow\}}'' again after some more editing.\\[1em]
\textbf{Note 2:} the ``\texttt{-narrow}'' versions of the bibliography style use the font ``PTSansNarrow-TLF'' for typesetting the DOIs in a compact way. This font needs to be available on your \LaTeX\ system. It is part of the \href{https://www.ctan.org/pkg/paratype}{``paratype'' package}, and many distributions (such as MikTeX) have it automatically installed. If you do not have this package yet and want to use a ``\texttt{-narrow}'' bibliography style then use your \LaTeX\ system's package installer to add it. If this is not possible you can also revert to the respective bibliography styles without the ``\texttt{-narrow}'' in the file name.\\[1em]
DVI-based processes to compile the template apparently cannot handle the different font so, by default, the template file uses the \texttt{abbrv-doi} bibliography style but the compiled PDF shows you the effect of the \texttt{abbrv-doi-hyperref-narrow} style.
\end{itemize}

\section{Bibliography Instructions}

\begin{itemize}
\item Sort all bibliographic entries alphabetically but the last name of the first author. This \LaTeX/bib\TeX\ template takes care of this sorting automatically.
\item Merge multiple references into one; e.\,g., use \cite{Max:1995:OMF,Kitware:2003} (not \cite{Kitware:2003}\cite{Max:1995:OMF}). Within each set of multiple references, the references should be sorted in ascending order. This \LaTeX/bib\TeX\ template takes care of both the merging and the sorting automatically.
\item Verify all data obtained from digital libraries, even ACM's DL and IEEE Xplore  etc.\ are sometimes wrong or incomplete.
\item Do not trust bibliographic data from other services such as Mendeley.com, Google Scholar, or similar; these are even more likely to be incorrect or incomplete.
\item Articles in journal---items to include:
  \begin{itemize}
  \item author names
	\item title
	\item journal name
	\item year
	\item volume
	\item number
	\item month of publication as variable name (i.\,e., \{jan\} for January, etc.; month ranges using \{jan \#\{/\}\# feb\} or \{jan \#\{-{}-\}\# feb\})
  \end{itemize}
\item use journal names in proper style: correct: ``IEEE Transactions on Visualization and Computer Graphics'', incorrect: ``Visualization and Computer Graphics, IEEE Transactions on''
\item Papers in proceedings---items to include:
  \begin{itemize}
  \item author names
	\item title
	\item abbreviated proceedings name: e.\,g., ``Proc.\textbackslash{} CONF\_ACRONYNM'' without the year; example: ``Proc.\textbackslash{} CHI'', ``Proc.\textbackslash{} 3DUI'', ``Proc.\textbackslash{} Eurographics'', ``Proc.\textbackslash{} EuroVis''
	\item year
	\item publisher
	\item town with country of publisher (the town can be abbreviated for well-known towns such as New York or Berlin)
  \end{itemize}
\item article/paper title convention: refrain from using curly brackets, except for acronyms/proper names/words following dashes/question marks etc.; example:
\begin{itemize}
	\item paper ``Marching Cubes: A High Resolution 3D Surface Construction Algorithm''
	\item should be entered as ``\{M\}arching \{C\}ubes: A High Resolution \{3D\} Surface Construction Algorithm'' or  ``\{M\}arching \{C\}ubes: A high resolution \{3D\} surface construction algorithm''
	\item will be typeset as ``Marching Cubes: A high resolution 3D surface construction algorithm''
\end{itemize}
\item for all entries
\begin{itemize}
	\item DOI can be entered in the DOI field as plain DOI number or as DOI url; alternative: a url in the URL field
	\item provide full page ranges AA-{}-BB
\end{itemize}
\item when citing references, do not use the reference as a sentence object; e.\,g., wrong: ``In \cite{Lorensen:1987:MCA} the authors describe \dots'', correct: ``Lorensen and Cline \cite{Lorensen:1987:MCA} describe \dots''
\end{itemize}

\section{Example Section}

Lorem\marginpar{\small You can use the margins for comments while editing the submission, but please remove the marginpar comments for submission.} ipsum dolor sit amet, consetetur sadipscing elitr, sed diam
nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat,
sed diam voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem
ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur
sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et
dolore magna aliquyam erat, sed diam
voluptua~\cite{Kitware:2003,Max:1995:OMF}. At vero eos et accusam et
justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit
amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor
invidunt ut labore et dolore magna aliquyam erat, sed diam
voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est.

\section{Exposition}

Duis autem vel eum iriure dolor in hendrerit in vulputate velit esse
molestie consequat, vel illum dolore eu feugiat nulla facilisis at
vero eros et accumsan et iusto odio dignissim qui blandit praesent
luptatum zzril delenit augue duis dolore te feugait nulla
facilisi. Lorem ipsum dolor sit amet, consectetuer adipiscing elit,
sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna
aliquam erat volutpat~\cite{Kindlmann:1999:SAG}.

\begin{equation}
\sum_{j=1}^{z} j = \frac{z(z+1)}{2}
\end{equation}

Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam
nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat,
sed diam voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem
ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur
sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et
dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam
et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
takimata sanctus est Lorem ipsum dolor sit amet.

\subsection{Lorem ipsum}

Lorem ipsum dolor sit amet (see \autoref{tab:vis_papers}), consetetur sadipscing elitr, sed diam
nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat,
sed diam voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem
ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur
sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et
dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam
et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit
amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor
invidunt ut labore et dolore magna aliquyam erat, sed diam
voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. 

\begin{table}[tb]
  \caption{VIS/VisWeek accepted/presented papers: 1990--2016.}
  \label{tab:vis_papers}
  \scriptsize%
	\centering%
  \begin{tabu}{%
	r%
	*{7}{c}%
	*{2}{r}%
	}
  \toprule
   year & \rotatebox{90}{Vis/SciVis} &   \rotatebox{90}{SciVis conf} &   \rotatebox{90}{InfoVis} &   \rotatebox{90}{VAST} &   \rotatebox{90}{VAST conf} &   \rotatebox{90}{TVCG @ VIS} &   \rotatebox{90}{CG\&A @ VIS} &   \rotatebox{90}{VIS/VisWeek} \rotatebox{90}{incl. TVCG/CG\&A}   &   \rotatebox{90}{VIS/VisWeek} \rotatebox{90}{w/o TVCG/CG\&A}   \\
  \midrule
	2016 & 30 &   & 37 & 33 & 15 & 23 & 10 & 148 & 115 \\
  2015 & 33 & 9 & 38 & 33 & 14 & 17 & 15 & 159 & 127 \\
  2014 & 34 &   & 45 & 33 & 21 & 20 &   & 153 & 133 \\
  2013 & 31 &   & 38 & 32 &   & 20 &   & 121 & 101 \\
  2012 & 42 &   & 44 & 30 &   & 23 &   & 139 & 116 \\
  2011 & 49 &   & 44 & 26 &   & 20 &   & 139 & 119 \\
  2010 & 48 &   & 35 & 26 &   &   &   & 109 & 109 \\
  2009 & 54 &   & 37 & 26 &   &   &   & 117 & 117 \\
  2008 & 50 &   & 28 & 21 &   &   &   & 99 & 99 \\
  2007 & 56 &   & 27 & 24 &   &   &   & 107 & 107 \\
  2006 & 63 &   & 24 & 26 &   &   &   & 113 & 113 \\
  2005 & 88 &   & 31 &   &   &   &   & 119 & 119 \\
  2004 & 70 &   & 27 &   &   &   &   & 97 & 97 \\
  2003 & 74 &   & 29 &   &   &   &   & 103 & 103 \\
  2002 & 78 &   & 23 &   &   &   &   & 101 & 101 \\
  2001 & 74 &   & 22 &   &   &   &   & 96 & 96 \\
  2000 & 73 &   & 20 &   &   &   &   & 93 & 93 \\
  1999 & 69 &   & 19 &   &   &   &   & 88 & 88 \\
  1998 & 72 &   & 18 &   &   &   &   & 90 & 90 \\
  1997 & 72 &   & 16 &   &   &   &   & 88 & 88 \\
  1996 & 65 &   & 12 &   &   &   &   & 77 & 77 \\
  1995 & 56 &   & 18 &   &   &   &   & 74 & 74 \\
  1994 & 53 &   &   &   &   &   &   & 53 & 53 \\
  1993 & 55 &   &   &   &   &   &   & 55 & 55 \\
  1992 & 53 &   &   &   &   &   &   & 53 & 53 \\
  1991 & 50 &   &   &   &   &   &   & 50 & 50 \\
  1990 & 53 &   &   &   &   &   &   & 53 & 53 \\
  \midrule
  \textbf{sum} & \textbf{1545} & \textbf{9} & \textbf{632} & \textbf{310} & \textbf{50} & \textbf{123} & \textbf{25} & \textbf{2694} & \textbf{2546} \\
  \bottomrule
  \end{tabu}%
\end{table}

\subsection{Mezcal Head}

Lorem ipsum dolor sit amet (see \autoref{fig:sample}), consetetur sadipscing elitr, sed diam
nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat,
sed diam voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem
ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur
sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et
dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam
et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
takimata sanctus est Lorem ipsum dolor sit amet. 

\subsubsection{Duis Autem}

Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam
nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat,
sed diam voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem
ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur
sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et
dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam
et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit
amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor
invidunt ut labore et dolore magna aliquyam erat, sed diam
voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est. Lorem
ipsum dolor sit amet.

\begin{figure}[tb]
 \centering
 \includegraphics[width=\columnwidth]{paper-count-w-2015-new}
 \caption{A visualization of the 1990--2015 data from \autoref{tab:vis_papers}. The image is from \cite{Isenberg:2017:VMC} and is in the public domain.}
 \label{fig:sample}
\end{figure}

\subsubsection{Ejector Seat Reservation}

Duis autem~\cite{Lorensen:1987:MCA}\footnote{The algorithm behind
Marching Cubes \cite{Lorensen:1987:MCA} had already been
described by Wyvill et al. \cite{Wyvill:1986:DSS} a year
earlier.} vel eum iriure dolor in hendrerit
in vulputate velit esse molestie consequat,\footnote{Footnotes
appear at the bottom of the column.} vel illum dolore eu
feugiat nulla facilisis at vero eros et accumsan et iusto odio
dignissim qui blandit praesent luptatum zzril delenit augue duis
dolore te feugait nulla facilisi. Lorem ipsum dolor sit amet,
consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt
ut laoreet dolore magna aliquam erat volutpat.


\paragraph{Confirmed Ejector Seat Reservation}

Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper
suscipit lobortis nisl ut aliquip ex ea commodo
consequat~\cite{Nielson:1991:TAD}. Duis autem vel eum iriure dolor in
hendrerit in vulputate velit esse molestie consequat, vel illum dolore
eu feugiat nulla facilisis at vero eros et accumsan et iusto odio
dignissim qui blandit praesent luptatum zzril delenit augue duis
dolore te feugait nulla facilisi.

\paragraph{Rejected Ejector Seat Reservation}

Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper
suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem
vel eum iriure dolor in hendrerit in vulputate velit esse molestie


\section{Conclusion}

Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam
nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat,
sed diam voluptua. At vero eos et accusam et justo duo dolores et ea
rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem
ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur
sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et
dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam
et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea
takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit
amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor
invidunt ut labore et dolore magna aliquyam erat, sed diam
voluptua. At vero eos et accusam et justo duo dolores et ea
rebum.


\acknowledgments{
The authors wish to thank A, B, and C. This work was supported in part by
a grant from XYZ.}

\bibliographystyle{abbrv-doi}

