Find research datasets worth reusing

Figure 4 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 4 - A map of the social network of a single participant who contributed about 500 observations to iSpotnature.org at the locations shown by blue dots, with other iSpot participants, living at locations shown by red stars, who provided determinations for those observations. The location of the observation and the location of the person identifying it are joined by a line.

Figure 2 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 2 - The conceptual social network structure of iSpot, showing the group (3 of the 8) compartmentalization and its interaction. Not shown is the learner-mentor interaction within each group. (Icons http://www.icons-land.com)

Figure 9 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 9 - For ispotnature.org in the 4-year period shown: a The percentage of observations submitted each month that received a likely ID and the percentage where an expert provided or agreed the determination b The number of observations submitted per month.

Figure 10 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 10 - The cumulative frequency distribution for the time taken for a likely ID to be acquired by observations submitted to ispotnature.org without an organism name. n = 100,703 observations.

Figure 1 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 1 - Schematic of three network structures, from equal status, to expert status and a hybrid system. (Icons http://www.icons-land.com)

Figure 3 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 3 - The network linking participants who posted observations to iSpotnature.org without an identification and those providing a likely identification for those observations. The diagram is based on the sample of 5,000 identifications made up until 1 July 2014. This activity occurred over 32 days. The network contains 1,110 nodes linked by 2,876 edges. Red nodes (83.23%) are participants who only received identifications, green nodes (6.38%) are participants who only made identifications and blue nodes (10.39%) are participants who both made and received identifications.

Figure 5 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 5 - An example of an iSpot participant's profile showing their reputation and activity in the 8 groups.

Figure 8 from: Silvertown J, Harvey M, Greenwood R, Dodd M, Rosewell J, Rebelo T, Ansine J, McConway K (2015) Crowdsourcing the identification of organisms: A case-study of iSpot. ZooKeys 480: 125-146. https://doi.org/10.3897/zookeys.480.8803

Figure 8 - The global distribution of observations made on ispotnature.org and ispot.org.za up to December 2013.

Figure 9 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 9 - Timeline. We will prepare and update the Web app (Aim 2) as we develop it for use in annotating gold standard audio data (Aim 1) and as we get feedback on its use in connection with Amazon's Mechanical Turk (Aim 2). Year 2 will consist primarily of testing the aggregation of annotated audio data for further analysis (Aim 2), to train an automated approach (Exploratory Aim), and to publish and present our findings.

Figure 2 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 2 - Mockup of audio recording annotation tool – Step 1: Selection. This figure shows a mockup of what an audio annotation Web application tool could look like. In this first step, (A) the Worker presses the Play icon to listen to the voice recording, (B) selects a problematic segment by clicking and dragging the mouse over the waveform, and (C) replays the recording if necessary and selects other problematic segments.

Figure 5 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 5 - DARPA-funded seedling project. This schematic represents our DARPA-funded seedling project to assess the feasibility of collecting phone voice recordings from PD patients for use in a competition.

Figure 4 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 4 - Audio recording annotation tool – Step 3: Rating. Following Figures 2 and 3, here the Worker rates how serious the problem is that is affecting the highlighted segment of the recording. In this example, the Worker indicates that the background noise (wind) is not good, but that it doesn't interfere with his/her ability to hear the voice in the recording.

Figure 7 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 7 - Example mPower patient voice data. In the mPower app, PD patients are prompted to perform the voice activity three times per day: once before taking their medication, a second time when they feel they are at their best after taking their medication, and a third "random" time. This figure shows example voice data for a single patient on medication (top) and at a "random" time, very likely off medication (bottom). On the left are waveforms, showing the acoustic voice signal over time (0-10 seconds), from which one can clearly see that the patient's voice trailed off to a minimum (bottom left) compared to after medication (top left). On the right are spectrograms, representing signal amplitude at different frequencies (0-5 kHz) over time (0-10 seconds). The spectrogram after medication (top right) has more uniform frequency bands across the recording compared to the rather "muddled" spectrogram recorded at the random time (bottom right).

Figure 3 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 3 - Audio recording annotation tool – Step 2: Annotation. Following Figure 2, here the Worker selects one or more categories describing why the highlighted segment in the audio waveform is problematic. In this example, there was a lot of background noise (wind).

Figure 6 from: Klein A (2016) Crowdsourcing voice editing and quality assessment of data collected from the largest mobile phone-based research study of Parkinson disease. Research Ideas and Outcomes 2: e8848. https://doi.org/10.3897/rio.2.e8848

Figure 6 - Android and iOS Parkinson app screenshots. Top: Android PD app screenshots showing instructions for the phonation (voice) task. Bottom: mPower PD app screenshots. Each participant in the mPower study is prompted to perform a voice activity three times a day. The rightmost screenshot demonstrates the visual feedback that is provided during audio recording, to try to keep the voice at the best amplitude for recording.

Figure 7 from: Klein A (2016) A game for crowdsourcing the segmentation of BigBrain data. Research Ideas and Outcomes 2: e8816. https://doi.org/10.3897/rio.2.e8816

Figure 7 - Timeline. We will prepare the gold standard evaluation data (Aim 1) and develop and update the game (Aim 2) through the 18th month as we get feedback on its use. Year 2 will consist primarily of testing the aggregation of segmented data (Aim 3), exploring how well the game can generalize to segmentation of every region of the BigBrain (Exploratory Aim 1), and training and testing an automated approach that learns from the crowdsourced data (Exploratory Aim 2), as well as to publish and present our findings.

Figure 6 from: Klein A (2016) A game for crowdsourcing the segmentation of BigBrain data. Research Ideas and Outcomes 2: e8816. https://doi.org/10.3897/rio.2.e8816

Figure 6 - Example joint fusion of multiple label assignments. This macroscopic brain atlas was built from 20 individually labeled atlases, using joint fusion (Wang and Yushkevich 2013), after nonlinearly registering the 20 to a template (http://mindboggle.info/data.html).

Figure 2 from: Klein A (2016) A game for crowdsourcing the segmentation of BigBrain data. Research Ideas and Outcomes 2: e8816. https://doi.org/10.3897/rio.2.e8816

Figure 2 - Example hippocampal subfield labels. Left: Example of hippocampal subfield labeling. Right: The first work demonstrating hippocampal subfield labels in MRI space that are derived from ground-truth histological imaging (Adler et al. 2014).

Figure 5 from: Klein A (2016) A game for crowdsourcing the segmentation of BigBrain data. Research Ideas and Outcomes 2: e8816. https://doi.org/10.3897/rio.2.e8816

Figure 5 - Example boundary estimates in the BigBrain hippocampus. This figure shows example color line drawings atop a (low-resolution) image tile containing a small portion of the BigBrain's right hippocampus. The lines represent a novice's estimates of the CA1/subiculum cytoarchitectonic boundary. This boundary is extremely difficult, as it results in the greatest overall disagreement among hippocampal subfield labeling protocols (Yushkevich et al. 2015).