Good morning ladies and gentlemen, I am Jun Haeng Lee, Samsung Medical Center, Seoul, Korea. My topic today is ‘Table and figure in the manuscript: how to prepare’. 

Determining the quality of a digital image file is not as simple as that of a printed image. In some cases, an image may appear to be of good quality when viewed on a computer monitor, but the quality is far below that needed for the publication. After my presentation, you can have some basic understanding of digital images required for manuscript preparation. 
+++++
Let’s start with tables. 
What can be the most common error in the table? As you can see, any article is made of text and tables. In some manuscript, the information is simply duplicated in the text and the table. /// We need to save the space in the paper and save the Earth. Only most important data from the table need to be mentioned in the text. 
What can be the most easy way of making tables? This is a table in the journal Endoscopy. 
It is generally recommended not to use Excel or comparable spreadsheet program for making tables. Most author guidelines recommend to created tables using the table formatting and editing feature of word processing software. However, in my experience, it is very easy to make a small mistake in entering the numbers in the word processing software. I am still using Excel spreadsheet without making any trouble in manuscript preparation and publication. /// In my opinion, both spreadsheet and word processors are OK. But make the table once and then use copy and paste function in order to avoid typing errors. 
Giving a title to each table is so important to attract reader’s attention. If the title is too short, it is difficult to get what’s the table is talking about. If the title is too long, the readers may turn to the next page.  
+++++
From now on, I’d like to find with you what’s wrong with the pictures.
This is a figure of an article from my institution in the year 2015. We have done a lot of endoscopic submucosal dissection for early gastric cancers. Among more than 1,000 cases, there were 2 extragastric recurrences. And these are endoscopic, radiographic and pathological images of the two cases. Regarding today’s topic, did you find what’s wrong with this pictures? /// You can see the date of the examination and the partial name of the patients. The patient information should be removed before submission to the Journal.  
Every PACS system has a function of removing the patients’ annotation before saving the images to your computer.  Look at the right upper corner. /// I will remove the annotation. 
What’s wrong with this pictures? EMR-L was done with transparent cap at the distal end of the endoscope. The cap should be round. But the shape of the cap is a little bit different between the images. The vertical and horizontal ratio, aspect ratio, should not be changed in any science pictures. 
In some endoscopy centers, the standard aspect ratio was not followed. In that situation, the basic setting of the PACS system should be corrected. 
What’s wrong with this graph? At the first sight, it looks find. /// But in the closed up view, you can find that the resolution of the text is not high enough. 
What’s wrong with this images? The topic of this article is minimal change lesions that can be seen in patients with reflux symptoms. There is no ulcer or no erosion but subtle color changes or blurring of the squamocolumnar junction. So, the pictures should be color images. /// Black and white images can be acceptable in some situations, but most endoscopic pictures needs to be color images. 
OK. It should be like this. 
Did you find anything wrong with this article? This report from Asan Medical Center has very rare information about the natural course of non-curative resection without additional treatment. 
There were 3 mortality cases. However, there were no endoscopic pictures for those three cases. Endoscopic pictures of the 3 mortality cases would have been very informative.
This is another example of the same mistake. Endoscopy report without endoscopy pictures. 
What’s wrong with this pictures? You are right. The control lanes of experiment A and experiment B are exactly the same. It’s impossible. It’s a typical example of duplication.
Duplication and other violations of publication ethics should be screened in every review processes. 
+++++
Next topic is resolution.
In the medical literature there are 4 types of images. (1) statistical graphs, charts, and simple diagrams, (2) Photographic images such color photos, radiographs, ultrasound images, CT scans, MRI scans, electron micrographs, and photomicrographs, (3) illustrations, and finally video materials. All kinds of images require minimal resolutions for publication.
What is the resolution? Resolution refers to a measurement of clarity or detail of the displayed image and is expressed as the number of pixels displayed per unit length. The resolution of a printed image is most commonly expressed in terms of dots per inch (dpi) or pixels per inch (ppi). The most commonly recommended resolution for printing on paper depends on the nature of the images: 1) 300 dpi for color pictures, 2) 300 to 600 dpi for black and white pictures, 3) 600 to 900 dpi for combination art (photo and text), and 4) 900 to 1,200 dpi for line art.
Which image has higher resolution? I mean which image is more clear? Of course, left hand side is more clear. It mean left hand side image has more dots. 
If the size of printed image is the same, more dots, more pixels, more picture elements means higher quality, higher resolution. 
The resolution is usually represented as DPI. Dots per inch. In the digital image file, the number of pixels per inch itself has no meaning in terms of the image quality. For instance, if a color picture image file with 600 dpi is printed as a 1-inch picture, the image quality will certainly be good. However, if the same image file of 600 dpi is printed as a 10-inch picture, the final image quality will be much poorer. In a digital image file, the image “quality” (that is, the information density) is not determined by the number of pixels per inch but the number of actual pixels. 
Because a digital image file has no inherent physical dimensions, the number of pixels per inch itself has no meaning in terms of the image quality. 
The actual number of pixels is the key determinant of the image quality for a digital image. In this case, what is important is the pixel number, that is 1,000 pixels. The number of DPI, 250, has nothing to do with image quality.  
+++++
Next topic is vector image. 
This is a beautiful tulip and the image quality is so nice. Clean and smooth. 
However, in the close up view, you can find that the picture was composed pixels, picture elements. 
There are two methods of making a line in a digital image. Bitmap and vector. Pictures taken with a digital camera is a typical example of bitmap or raster images. 
A raster image, also called a bitmap image, is a collection of pixels (picture elements) arranged on a rectangular grid. Usually, more pixels mean higher image quality. Raster images are resolution-dependent, meaning they cannot be scaled up to an arbitrary resolution without loss of quality. The original raster image is always the most detailed image. 
Popular graphic formats for raster images are GIF, JPEG, PNG, and TIFF. The most popular image editor for raster images is Photoshop. 
A vector image is based on vectors. Because pixels do not exist in a vector image, it can be scaled up without loss of image quality. Popular graphic formats for vector images are EPS, SVG, and AI. The most popular image editor for vector images is Adobe Illustrator. 
The PDF is unique that it can handle both raster images and vector images.
How can vector images and raster images be distinguished? The simple way is to magnify the image to 500% or more. When raster images are scaled up, you can see a loss of image quality. On the other hand, vector images can be scaled up without loss of quality. 
Most formats that we use every day are bitmap (=rastor) formats. 
However, some journals, like lancet, may require vector images. 
From now on, I’d like to show you how to make a vector image file using the Prism software. WMF windows metafile is the only vector format in the Prism. You can export your drawing with WMF format and that’s it. 
The file size is extremely small. Just 5 kilo-bytes. Not 5 mega-bytes. 
When you scale up the image, the quality is always the same. 
The image quality is always perfect because it is a line rather than a dot. 
Using a graphic program, WMF can be easily transformed into an Adobe illustrator (.ai) file.
+++++
Next topic is pixels for publication. In order to determine the number of pixels, the physical size of the image must first be determined.
However, authors cannot know the final size of their image during preparation of images for submission. 
Only a few journals recommend the size of the artwork in their author guidelines. For example, lancet recommends 107mm width and 300 dpi. But this is a very rare example. 
In my opinion, universal recommendation could help authors prepare their images. The standard figure size of most academic journals is a single column. It is usually 3.5 inch or less. And the standard DPI for line art is 900. Therefore, an image file of 900 dpi and 4 inches is of sufficient quality for most publications; it means 3,600 pixels in a horizontal line. It is recommended that authors use this number as a universal guide.
Once again, the standard figure size of most academic journals is a single column. It is usually 3.5 inch or less. And the standard DPI for line art is 900. Therefore, an image file of 900 dpi and 4 inches is of sufficient quality for most publications; it means 3,600 pixels in a horizontal line. 
+++++
My last topic is making a TIFF file from a powerpoint file. 
The most common error encountered when preparing images with PowerPoint involves the “Save As” function. If TIFF or JPEG image files are created directly from a PowerPoint file using the “Save As” function. 
It typically produces a low-resolution image, such as a 960× 720 pixel image. These low-quality images cannot be used for publication purposes. 
How can we make a high quality TIFF from a PPT file. I recommend to use PDF as an interim format, because a PDF file can handle both raster images and vector images. This can be accomplished by the following steps: first, create a high-quality image using PowerPoint. Second, convert the PowerPoint file into a PDF file using the “Save As” function. Third, open the PDF file with a sufficient resolution using Photoshop. This step is called “rasterization.” Fourth, convert the image to a high-quality TIFF file using the “Save As” function of Photoshop. This image file should be acceptable for medical publication.
Step 2. Converting the PowerPoint file into a PDF file can be done with Acrobat or “Save As” function of Powerpoint. 
Step 3 means opening the PDF file with sufficient resolution using Photoshop. 
Saving as a TIFF is very easy. 
That’s it. Powerpoint was very useful for me to make a publication quality images. 
+++++
Before closing my talk, I’d like to say that the current author guideline of clinical endoscopy is much too simple. 
Regarding the figure, there are only two sentences. The recommended formats and resolution is still a classic paper print style. It needs to be updated as soon as possible. Thank your for your attention.