You are here

ImageCLEFmed: Medical classification

Primary tabs

Medical compound figure separation and multi-label classification task

The objective of this task is to work on compound figures of the biomedical literature and to separate them if possible and/or attach to the sub parts labels about the content.

                

Schedule

  • 12.11.2014: registration opens for all ImageCLEF tasks (until 30.04.2015)
  • 15.03.2015: data release
  • 01.05.2015: deadline for submission of runs by the participants
  • 15.05.2015: release of processed results by the task organizers
  • 07.06.2015: deadline for submission of working notes papers by the participants
  • 30.06.2015: notification of acceptance of the working notes papers
  • 15.07.2015: camera ready working notes papers
  • 08.-11.09.2015: CLEF 2015, Toulouse, France

Citations

  • When referring to ImageCLEFmed 2015 task general goals, general results, etc. please cite the following publication:
    • Alba García Seco de Herrera, Henning Müller and Stefano Bromuri , Overview of the ImageCLEF 2015 medical classification task, in: CLEF working notes 2015, Toulouse, France, 2015.
    • BibText:
      @InProceedings{GMB2015,
        Title = {Overview of the {ImageCLEF} 2015 medical classification task},
        Author = {Garc\'ia Seco de Herrera, Alba and M\"uller, Henning and Bromuri, Stefano},
        Booktitle = {Working Notes of {CLEF} 2015 (Cross Language Evaluation Forum)},
        Year = {2015},
        Month = {September},
        Location = {Toulouse, France}
      }
  • When referring to ImageCLEFmed task in general, please cite the following publication:
    • Jayashree Kalpathy-Cramer, Alba García Seco de Herrera, Dina Demner-Fushman, Sameer Antani, Steven Bedrick and Henning Müller, Evaluating Performance of Biomedical Image Retrieval Systems –an Overview of the Medical Image Retrieval task at ImageCLEF 2004-2014 (2014), in: Computerized Medical Imaging and Graphics
    • BibText:
      @Article{KGD2014,
        Title = {Evaluating Performance of Biomedical Image Retrieval Systems-- an Overview of the Medical Image Retrieval task at {ImageCLEF} 2004--2014},
        Author = {Kalpathy--Cramer, Jayashree and Garc\'ia Seco de Herrera, Alba and Demner--Fushman, Dina and Antani, Sameer and Bedrick, Steven and M\"uller, Henning},
        Journal = {Computerized Medical Imaging and Graphics},
        Year = {2014}
      }

Motivation

An estimated 40% of the figures in PubMed Central are compound figures (images consisting of several sub figures) like the images above. When data of articles is made available digitally, often the compound images are not separated but made available in a single block. Information retrieval systems for images should be capable of distinguishing the parts of compound figures that are relevant to a given query. A major step for making the content of the compound figures accessible is the detection of compound figures and then their separation into sub figures that can subsequently be classified into modalities and made available for research. The medical classification task of ImageCLEF 2015 uses a subset of PubMed Central.

Task overview

There are four types of tasks in 2015:
  • Compound figure detection: Compound figure identification is therefore a required first step to make available compound images from the literature. Therefore, the goal of this task is to identify whether a figure is a compound figure or not. The task makes training data available containing compound and non compound figures from the biomedical literature.
  • Multi-label classification: Characterization of compound figures is difficult, as they may contain subfigures from various imaging modalities or image types. This task aims to label each compound figure with each of the modalities (of the 30 classes of a defined hierarchy shown below) of the subfigures contained without knowing where the separation lines are.
  • Figure separation: This task was first introduced in 2013. The task makes available training data with separation labels of the figures and then a test data set where the labels will be made available after the submission of the results. In 2015, a larger number of compound figures is distributed compared to the previous task.
  • Subfigure classification: Similar to the modality classification task organized in 2011-2013 this task aims to classify images into the 30 classes of the hierarchy shown below. The images are the subfigures extracted from the compound figures distributed for the figure separation task.

Classification hierarchy

The following hierarchy is used for the modality classification, same classes as in ImageCLEF 2012-2013, although this year '[COMP] Compound or multiplane images' is not a class.

Class codes with descriptions (class codes need to be specified in run files):
([Class code] Description)

  • [Dxxx] Diagnostic images:
    • [DRxx] Radiology (7 categories):
      • [DRUS] Ultrasound
      • [DRMR] Magnetic Resonance
      • [DRCT] Computerized Tomography
      • [DRXR] X-Ray, 2D Radiography
      • [DRAN] Angiography
      • [DRPE] PET
      • [DRCO] Combined modalities in one image
    • [DVxx] Visible light photography (3 categories):
      • [DVDM] Dermatology, skin
      • [DVEN] Endoscopy
      • [DVOR] Other organs
    • [DSxx] Printed signals, waves (3 categories):
      • [DSEE] Electroencephalography
      • [DSEC] Electrocardiography
      • [DSEM] Electromyography
    • [DMxx] Microscopy (4 categories):
      • [DMLI] Light microscopy
      • [DMEL] Electron microscopy
      • [DMTR] Transmission microscopy
      • [DMFL] Fluorescence microscopy
    • [D3DR] 3D reconstructions (1 category)
  • [Gxxx] Generic biomedical illustrations (12 categories):
    • [GTAB] Tables and forms
    • [GPLI] Program listing
    • [GFIG] Statistical figures, graphs, charts
    • [GSCR] Screenshots
    • [GFLO] Flowcharts
    • [GSYS] System overviews
    • [GGEN] Gene sequence
    • [GGEL] Chromatography, Gel
    • [GCHE] Chemical structure
    • [GMAT] Mathematics, formulae
    • [GNCP] Non-clinical photos
    • [GHDR] Hand-drawn sketches

Data collection

The dataset used in this task is a subset of images contained in articles from the biomedical literature extracted from the PubMed Central.

  • Compound figure detection:
    20,000 images labelled as compound figures or not compound figures are distributed.
  • Figure separation:
    A subset of the compound figures from the compound figure detection task is distributed to be separated into subfigures.
  • Multi-label classification:
    A subset of the compound figures are distributed for the multi-label task.
  • Subfigure classification:
    Figures from the multi-label classification task are separated into subfigures and each of the subfigures are labelled.
    If the figure ID is "1297-9686-42-10-3", then the corresponding subfigures IDs are "1297-9686-42-10-3-1", "1297-9686-42-10-3-2", "1297-9686-42-10-3-3" and "1297-9686-42-10-3-4"

Submission instructions

Compound figure detection

The format of the result submission for the Compound figure detection subtask should be the following:

1471-2458-10-S1-S4-3 COMP 0.9
1471-2458-10-52-5 COMP 1
1471-2458-11-133-1 COMP 0.4
1423-0127-17-34-8 NOCOMP 0.8
1465-9921-6-21-6 NOCOMP 0.9
...

where:

  • The first column contains the figure-ID (IRI). This ID does not contain the file format ending and it should not represent a file path.
  • The second column is the classcode.
  • The third column represents the normalized score (between 0 and 1) that your system assigned to the specific result.

You need to respect the following constraints:

  • Each specified figure has to be part of the collection (dataset).
  • A figure cannot be contained more than once.
  • All figures of the test set have to be contained in the runfile.
  • Only known class codes are accepted.

Please note that each group is allowed to submit a maximum of 10 runs.

Compound figure separation

The format of the result submission for the compound figure separation subtask has to be an XML file with the following structure :

<?xml version="1.0" encoding="UTF-8"?>
<annotations>
   <annotation>
      <filename>1423-0127-16-105-5</filename>
      <object>
         <point x="0" y="1" />
         <point x="299" y="1" />
         <point x="0" y="242" />
         <point x="299" y="242" />
      </object>
      <object>
         <point x="300" y="1" />
         <point x="600" y="1" />
         <point x="300" y="242" />
         <point x="600" y="242" />
      </object>
      ...
   </annotation>
   <annotation>
      <filename>1423-0127-16-22-1</filename>
      ...
   </annotation>
   ...
</annotations>
where:
  • The root element is <annotations>.
  • The root contains one <annotation> element per image. Each one of these elements has to contain :
    • A <filename> element with the name of the compound image (IRI) (excluding the file extension)
    • One or more <object> elements that define the bounding box of each subfigure in the image. Each <object> must contain :
      • 4 <point> elements that define the 4 corners of the bounding box. The <point> elements must have two attributes (x and y), which correspond to the horizontal and vertical pixel position, respectively. The preferred order of the points is :
        1. top-left
        2. top-right
        3. bottom-left
        4. bottom-right
You also need to respect the following constraints:
  • Each specified image has to be part of the collection (dataset).
  • An Image cannot appear more than once in a single XML results file.
  • All the images of the testset must be contained in the runfile.
  • The resulting XML file MUST validate against the XSD schema that will be provided.

Multi-label classification

The format of the result submission for the Multi-label classification classification subtask should be the following:

1751-0147-52-24-3 DRMR DRXR
1475-925X-6-10-8 DRUS GHDR
1471-2210-10-7-3 GCHE
1475-2875-6-10-2 GFIG
...

where:
  • The first column contains the figure-ID (IRI). This ID does not contain the file format ending and it should not represent a file path.
  • The rest of the columns contain the classcode.
You need to respect the following constraints:
  • Each specified figure has to be part of the collection (dataset).
  • A figure cannot be contained more than once.
  • All figures of the test set have to be contained in runfile.
  • Only known classcodes are accepted.
Please note that each group is allowed a maximum of 10 runs.

Subfigure classification

Similar to the compound figure detection subtask, the format of the result submission for the Subfigure classification subtask should be the following:

1743-422X-4-12-1-4 D3DR 0.9
1471-2156-8-36-3-4 GFIG 1
1475-2859-9-86-6-1 GFIG 0.4
1475-2840-10-59-2-4 DMLI 0.8
...

where:
  • The first column contains the subfigure-ID (IRI). This ID does not contain the file format ending and it should not represent a file path.
  • The second column is the classcode.
  • The third column represents the normalized score (between 0 and 1) that your system assigned to the specific result.
You need to respect the following constraints:
  • Each specified subfigure has to be part of the collection (dataset).
  • A subfigure cannot be contained more than once.
  • All subfigures of the test set have to be contained in runfile.
  • Only known class codes are accepted.
Please note that each group is allowed a maximum of 10 runs.

Evaluation methodology

Compound Figure Separation : The Java Archive (JAR) containing the application to run the evaluation can be downloaded below. The source code of the application is also contained in the ZIP archive.

Download Compound Figure Separation Evaluation Tool & Source

More details on the evaluation will be provided at a later time

Results

Compound figure detection
Group name Run Run type Correctly classified in %
Biomedical Computer Science Group task1_run2_mixed_sparse1 mixed 85.39
Biomedical Computer Science Group task1_run1_mixed_stemDict mixed 83.88
Biomedical Computer Science Group task1_run3_mixed_sparse2 mixed 80.07
Biomedical Computer Science Group task1_run4_mixed_bestComb mixed 78.32
Biomedical Computer Science Group task1_run6_textual_sparseDict textual 78.34
CIS UDEL exp1 visual 82.82
Biomedical Computer Science Group task1_run5_visual_sparseSift visual 72.51
Multi-label classification
Group name Run Hamming Loss
IIS output_6 0.0817
IIS output_8 0.0785
IIS output_9 0.0710
IIS output_7 0.0700
IIS output_10 0.0696
IIS output_5 0.0680
IIS output_1 0.0678
IIS output_3 0.0675
MindLAB predictions_Mindlab_ImageclefMed_multilabel_test_comb2lbl 0.0674
IIS output_4 0.0674
IIS output_2 0.0671
MindLAB predictions_Mindlab_ImageclefMed_multilabel_test_comb1lbl 0.0500
Figure separation
Group name Run Run type Correctly classified in %
NLM run2_whole visual 84.64
NLM run1_whole visual 79.85
AAUITEC aauitec_figsep_combined visual 49.40
AAUITEC aauitec_figsep_edge visual 35.48
AAUITEC aauitec_figsep_bond visual 30.22
Subfigure classification
Group name Run Run type Correctly classified in %
Biomedical Computer Science Group task4_run5_train_20152013.txt mixed 67.60
Biomedical Computer Science Group task4_run4_clean_rf.txt mixed 67.24
Biomedical Computer Science Group task4_run1_combination.txt mixed 66.48
Biomedical Computer Science Group task4_run8_clean_short_rf.txt mixed 66.44
Biomedical Computer Science Group task4_run7_clean_comb_librf.txt mixed 65.99
Biomedical Computer Science Group task4_run6_clean_libnorm.txt mixed 64.34
Biomedical Computer Science Group task4_run3_textual.txt textual 60.91
Biomedical Computer Science Group task4_run2_visual.txt visual 60.91
CMTECH resultsSubfigureRunWholeCov.txt visual 52.98
CMTECH resultsSubfigure.txt visual 48.61
BMET sf_run_3.txt visual 45.63
BMET sf_run_6.txt visual 45.00
BMET sf_run_4.txt visual 44.34
BMET sf_run_2.txt visual 43.62
BMET sf_run_1.txt visual 37.56
BMET sf_run_5.txt visual 37.56

Organizers

  • Alba García Seco de Herrera, National Library of Medicine (NLM/NIH), Bethesda, MD, USA, albagarcia(replace-by-an-at)nih.gov
  • Stefano Bromuri, University of Applied Sciences Western Switzerland in Sierre, Switzerland, stefano.bromuri(replace-by-an-at)hevs.ch

Acknowledgements