Matching Forensic Sketches to Mug Shot Photos
Summary (3 min read)
Introduction
- Coronary artery disease (CAD) causes about 45% of all deaths in Europe.
- Therefore, the main aim of exercisebased cardiac rehabilitation is to improve peak VO2.
- The most common training programmes currently used in cardiac rehabilitation are continuous training (CT) and interval training (IT) or a mixture.
- A decade ago, it was shown that IT, consisting of several bouts of exercise at a higher intensity (80–90% of the peak VO2) alternated by active recovery periods, may provide a safe alternative to CT in CAD patients.
Participants
- The authors study included 20 male CAD patients (mean age 62.4 6.1), referred to the Cardiac Rehabilitation Unit from the University Hospitals of Leuven between July 2014–November 2015.
- The study was approved by the local ethics committee (Commissie Medische Ethiek KU Leuven).
- After inclusion, two patients dropped out before any tests were performed: one patient because of an old knee injury and one due to severe arrhythmias at high training intensities.
- There were no significant differences between the baseline characteristics of the patients with or without the dropouts (data not shown).
Measurements
- Height (cm) and weight (kg) were measured, and body mass index (BMI) (kg/m2) was calculated by dividing weight (kg) by height squared (m2).
- Before starting the cardiac rehabilitation, subjects performed a maximal graded exercise test on a cycle ergometer (Oxycon Pro, Jaeger, CareFusion, Germany) supervised by a trained exercise physiologist.
- The peak HR was defined as the highest HR reached at the end of the test.
- The peak VO2 was determined as the at KU Leuven University Library on April 21, 2016cpr.sagepub.comDownloaded from VO2 during the last full bout of 30 s of the test.
- A peak respiratory exchange ratio of at least 1.10 defined a maximal effort.28 Individual peak VO2 results were compared with predicted reference values of Wasserman et al. to determine percentage of predicted peak VO2.
IT and CT familiarisation sessions
- After inclusion, subjects were randomly assigned on a 1:1 base to either training group 1 (starting with IT) or group 2 (starting with CT).
- They performed a total of six supervised IT (n¼ 3) and CT (n¼ 3) sessions on a cycle ergometer (Ergo-fit, Gymna, Brussels, Belgium), three times per week for two weeks.
- The IT and CT sessions were alternated and aimed to familiarise the subjects with both training programmes.
- A Garmin chest strap and wristwatch continuously monitored HR (Garmin, Garmin International, Kansas, USA), and training loads were adapted throughout the training to ensure that participants would remain within the prescribed HR zones.
IT and CT test sessions
- The two-week run-in period was followed by four test sessions within two weeks: the IT and CT according to the protocol of the SAINTEX-CAD study,25 which was based on Wisloff et al.
- An online randomisation procedure was performed to determine the sequence of the four test sessions.
- During the test sessions, HR was continuously monitored using a Polar chest strap (Polar, Polar Electro, Kempele, Finland).
- As beta-blocking medication might influence HR, the authors also asked their patients to take their medication at the same time each morning of the test.
- Breath by breath gas exchange measurements (VE, VO2, VCO2), averaged each 30 s, were provided by the Oxycon mobile device (Oxycon mobile Jaeger, CareFusion, Germany) and were used to calculate the EE (indirect calorimetry) according to the American College of Sports Medicine (ACSM) guidelines; one litre of O2 uptake equals 5 kcal.
Statistical analysis
- All values are expressed as mean standard deviation (SD), median and range, or as number and percentage.
- All data were normally distributed according to the Shapiro-Wilk test for normality.
- To examine the differences in EE, exercise intensity or lactate between the ITw, CTw, ITc and CTc session, a repeated measures analysis of variance was performed.
- A p-value< 0.05 was considered statistically significant.
Results
- A total of 18 patients completed the six training sessions and the four tests; baseline characteristics are presented in Table 1.
- Patients did not change medication during the study period.
- In Figure 4 and Table 2, the mean EE per training session is presented.
- The patients performed the tests within the prescribed HR zones (taking into account the end-interval values for the IT sessions) as shown in Table 2 and in Figure 5.
Discussion
- This is the first study to objectively measure EE during IT and CT.
- Therefore in previous studies, training sessions were designed to be isocaloric based on a theoretical calculation of their respective EE.
- The present study examines whether the IT and CT programmes described in the protocol of the SAINTEX-CAD study25 (based on a study of Wisloff et al.: ITw and CTw),22 and the actually achieved intensities in the SAINTEX-CAD study (ITc and CTc),10 were truly isocaloric by objectively measuring the EE.
- (a) a similar EE for the protocols of Wisloff et al, also known as The authors results showed.
- (ITw vs CTw); (b) a significantly higher EE for the CT of the SAINTEX-CAD study compared to the IT (ITc vs CTc); (c) a significant increase in lactate after ITw, ITc and CTc, but not Table 1.
Medication
- Nevertheless, most of the patients had difficulties maintaining this high HR zone and needed constant encouragement from the supervising exercise physiologist.
- According to Skinner’s three-phase model, training modalities below the first ventilatory threshold (60– 70% peak HR) do not exceed a 2mmol/l lactate level.
- When lactate levels exceed 4mmol/l, passing the second ventilatory threshold, the exercise is considered anaerobic (>90% peak HR).
Training intensity
- At KU Leuven University Library on April 21, 2016cpr.sagepub.comDownloaded from between the lactate levels of the CT and IT, with IT showing lactate values> 4mmol/l.
- This implies that the high intensity intervals may have been performed anaerobically for most of the individuals (ITw 14/18; ITc 13/18), but not in all, since this secondary threshold is highly individual.
- The frequently used term ‘aerobic’ interval training may thus be inappropriate and misleading, but further research is needed.
- The authors can however conclude that the participants remained in the aerobic zone for both CT sessions (CTw 2.45mmol/l and CTc 3.41mmol/l).
- From the lactate levels of the CTw session, the authors can conclude that an intensity set at 70–75% of peak HR is simply insufficient as a training stimulus.
Limitations
- The results of this study should be interpreted within the context of its limitations.
- The first limitation of their study consists of a relatively small sample size.
- The authors expect, however, these variations to be random across all subjects, which should, therefore, not significantly influence their final result.
- As the authors previously reported that substantial changes in peak VO2 at KU Leuven University Library on April 21, 2016cpr.sagepub.comDownloaded from and peak HR already occur after six weeks of training,10 this run-in period could have influenced the high VO2 levels achieved at prescribed HR.
- Fifth, the authors did not measure lactate levels during the maximal exercise test, hence they were unable to compare these individual maximal lactate levels to the lactate levels found during the tests.
Conclusion
- The authors found that CT according to the SAINTEX-CAD study expended significantly more energy than IT.
- As previous studies showed IT and CT to either yield similar improvement in peak VO2, or a larger improvements after IT compared to CT, the authors can conclude that IT is more efficient in improving peak VO2, producing a larger gain during a shorter training duration and at a lower energy cost.
- Since EE is not a goal on its own in cardiac rehabilitation, the main question remains which training modality is the most appropriate to increase peak VO2.
- The authors can suggest that IT and CT are equally effective, if CT sessions are performed at sufficiently high intensities.
- The authors stress the importance of objectively measuring actual caloric expenditure by indirect calorimetry in pilot studies, rather than using a general formula for setting up isocaloric exercise training programmes.
Did you find this useful? Give us your feedback
Citations
610 citations
Cites methods from "Matching Forensic Sketches to Mug S..."
...In [23], a local feature-based discriminant analysis method was proposed to match a forensic sketch and a mug shot photo, and also other effective features can be used such as [24]....
[...]
382 citations
358 citations
Additional excerpts
...In our implementation, s ¼ 16....
[...]
...In this paper, 1 ¼ 2 and 2 ¼ 4....
[...]
342 citations
338 citations
Cites background or methods from "Matching Forensic Sketches to Mug S..."
...[14] proposed local feature-based discriminant analysis (LFDA)....
[...]
...The second family of approaches [17, 15, 14] focused on the classification stage and tried to design advanced classifiers to reduce the modality gap between features extracted from photos and sketches....
[...]
...This application leads to a number of studies on this topic [26, 27, 28, 31, 9, 14, 6]....
[...]
...We test our method with the same configuration of training and testing splits as [31, 14]....
[...]
References
46,906 citations
14,245 citations
"Matching Forensic Sketches to Mug S..." refers methods in this paper
...Such an approach is similar to other methods proposed in the literature [14], [15], [16] of matching near-infrared images (NIR) to visible light images (VIS), where local binary pattern [17] feature descriptors are used to describe both NIR and VIS images....
[...]
7,057 citations
5,563 citations
"Matching Forensic Sketches to Mug S..." refers background in this paper
...Image-based features have been shown to be successful in face recognition, most notably with the use of local binary patterns [19]....
[...]
2,952 citations
"Matching Forensic Sketches to Mug S..." refers background in this paper
...[27], 2) 295 pairs from the XM2VTS database [28], and 3) 188 pairs...
[...]
Related Papers (5)
Frequently Asked Questions (8)
Q2. What is the proposed method for combining large feature size and small sample size?
In order to handle the combination of a large feature size and small sample size, an ensemble of linear discriminant classifiers called LFDA is proposed.
Q3. What other methods have been proposed to handle the SSS problem?
Other discriminant analysis methods have been proposed to handle the SSS problem, such as random sampling LDA [23], regularized LDA [24], and direct LDA [25].
Q4. Why are image descriptors not sufficiently verbose to describe a face?
Because most image descriptors are not sufficientlyverbose to fully describe a face image, the descriptors are computedover a set of uniformly distributed subregions of the face.
Q5. What is the main reason why the authors have not found a large number of forensic sketches?
the authors believe that with a larger number of forensic sketches, the authors could more properly train their discriminant and further improve the matching performance.
Q6. Why is the culprit being depicted in a forensic sketch considered a suspect?
This is because the culprit being depicted in a forensic sketch typically has committed a heinous crime (e.g., murder, rape, and armed robbery) that will receive a large amount of attention from investigators.
Q7. What is the approach to extract discriminant features?
A straightforward approach would be to apply classical subspace analysis (such as LDA) directly on , and to extract discriminant features for classification.
Q8. What are the key difficulties in matching forensic sketches?
The authors highlight two key difficulties in matching forensic sketches: 1) matching across image modalities and 2) performing face recognition despite possibly inaccurate depictions of the face.