Lameness detection in dairy cows: Part 1. How to distinguish between non-lame and lame cows based on differences in locomotion or behavior

Animals 2015,5, 838-860; doi:10.3390/ani5030387 OPEN ACCESS animals ISSN 2076-2615 www.mdpi.com/journal/animals Review Lameness Detection in Dairy Cows: Part 1. How to Distinguish between Non-Lame and Lame Cows Based on Differences in Locomotion or Behavior Annelies Van Nuffel 1,*, Ingrid Zwertvaegher 1, Liesbet Pluym 1, Stephanie Van Weyenberg 1, Vivi M. Thorup 2,3, Matti Pastell 4, Bart Sonck 5,6 and Wouter Saeys 7 1Technology and Food Science Unit–Precision Livestock Farming; The Institute for Agricultural and Fisheries Research (ILVO), Burgemeester van Gansberghelaan 115 bus 1, 9820 Merelbeke, Belgium; E-Mails: [email protected] (I.Z.); [email protected] (L.P.); [email protected] (S.V.W.) 2INRA, UMR 791 Systemic Modelling of Ruminant Nutrition, 16 rue Claude Bernard, 75231 Paris cedex 05, France; E-Mail: vi[email protected] 3AgroParisTech, UMR 791 Systemic Modelling of Ruminant Nutrition, 16 rue Claude Bernard, 75231 Paris cedex 05, France 4Natural Resources Institute Finland (Luke), Green Technology, Koetilantie 5, 00790 Helsinki, Finland; E-Mail: [email protected] 5Animal Sciences Unit, The Institute for Agricultural and Fisheries Research (ILVO), Scheldeweg 68, 9090 Melle, Belgium; E-Mail: [email protected] 6Department of Biosystems Engineering, Faculty of Bioscience Engineering, Ghent University, Coupure links 653, 9000 Gent, Belgium 7Division Mechatronics, Biostatistics and Sensors (MeBioS), Department of Biosystems, Katholieke Universiteit Leuven, Kasteelpark Arenberg 30 bus 2456, 3001 Heverlee, Belgium; E-Mail: [email protected] *Author to whom correspondence should be addressed; E-Mail: annelies.vannuf[email protected]. Academic Editor: Jon Huxley Received: 29 May 2015 / Accepted: 18 August 2015 / Published: 28 August 2015 Simple Summary: Scoring cattle for lameness based on changes in locomotion or behavior is essential for farmers to find and treat their lame animals. This review discusses the normal locomotion of cows in order to define abnormal locomotion due to lameness. It furthermore provides an overview of various relevant visual locomotion scoring systems Animals 2015,5839 that are currently being used as well as practical considerations when assessing lameness on a commercial farm. Abstract: Due to its detrimental effect on cow welfare, health and production, lameness in dairy cows has received quite a lot of attention in the last few decades—not only in terms of prevention and treatment of lameness but also in terms of detection, as early treatment might decrease the number of severely lame cows in the herds as well as decrease the direct and indirect costs associated with lameness cases. Generally, lame cows are detected by the herdsman, hoof trimmer or veterinarian based on abnormal locomotion, abnormal behavior or the presence of hoof lesions during routine trimming. In the scientific literature, several guidelines are proposed to detect lame cows based on visual interpretation of the locomotion of individual cows (i.e., locomotion scoring systems). Researchers and the industry have focused on automating such observations to support the farmer in finding the lame cows in their herds, but until now, such automated systems have rarely been used in commercial herds. This review starts with the description of normal locomotion of cows in order to define ‘abnormal’ locomotion caused by lameness. Cow locomotion (gait and posture) and behavioral features that change when a cow becomes lame are described and linked to the existing visual scoring systems. In addition, the lack of information of normal cow gait and a clear description of ‘abnormal’ gait are discussed. Finally, the different set-ups used during locomotion scoring and their influence on the resulting locomotion scores are evaluated. Keywords: lameness; dairy cattle; cow gait; behavior; visual locomotion scoring 1. Introduction The increasing demand for animal products has led to a rapid growth in livestock production, including dairy farming, during the last 20 years [1]. As a result, dairy farming systems worldwide have intensified, with more cattle on fewer farms and per caretaker and higher productivity per animal [2]. This trend reduces the farmer’s available time to observe and monitor the cows and jeopardizes the health of the cows, in particular the high-yielding ones. Lameness is considered to be the third most costly health problem of dairy cows, after reduced fertility and mastitis [3]. Nevertheless, lameness has not only been under-recorded on farms but its importance with regard to cow welfare, cow health and farm profitability has also been hugely underestimated [4], even though it induces both direct (drug treatment, veterinary costs and death) and indirect costs (reduced milk production, reproductive performance and life expectancy) [5]. With a lameness prevalence reaching up to 72% [6], the levels in dairy herds in Europe are unacceptably high. Therefore, minimizing the occurrence and impact of lameness is one of the greatest challenges the dairy industry is currently facing [7]. Unfortunately, many dairy farmers are unaware of the number of lame cows in their herd, and, if noticed, they often do not have enough time to treat them [8]. Because many farmers fail to notice or lack the time to look for lame cows, sensor technology is being used to develop automated lameness detection systems. Generally, lame cows are detected by the herdsman, hoof trimmer or veterinarian based on changes in Animals 2015,5840 cow gait, posture or behavior or the presence of hoof lesions during routine trimming. This review is the first of two focusing on cow characteristics that are used to visually detect lame cows in practice. These characteristics are grouped into those of cow gait, posture and general behavior. To know which alterations of these characteristics are due to lameness, information on the normal locomotion (gait and posture) and behavior of cows are firstly described. Next, the abnormal characteristics in cow gait, posture or behavior closely related to lameness are summarized. In addition, several lameness scoring systems are compared, and some considerations in applying them in practice are discussed. The second review focuses on sensor technology to automatically measure relevant cow characteristics of gait, posture or behavior to develop automated lameness detection systems that support farmers in identifying the lame cows in their herd [9]. 2. Definition of Lameness Lameness can be defined as the clinical manifestation of painful disorders, mainly related to the locomotor system, resulting in impaired movement or deviation from normal gait or posture. The severity of lameness can vary from stiffness or decreased symmetry of limb movement to an inability to bear weight on a limb, or even total recumbency [10,11]. Disorders can be located in the limbs or trunk of the animal, and may include painful lesions as well as mechanical defects resulting in physical incapability [12]. Gait problems are thus a manifestation of discomfort or pain. In dairy cattle, the main cause of lameness are claw lesions. Claw lesions can be divided into non-infectious (such as white line disease, sole ulcer, sole hemorhhage, interdigital hyperplasia) or infectious claw lesions (such as digital dermatitis, interdigital dermatitis, heel erosion, foot rot) [13]. In addition, lameness can also be associated with injury in the nervous system (such as obturator paralysis) as well as the musculoskeletal system (such as fractures, arthritis and tendonitis) [14]. How these different disorders specifically influence the gait of cows remains unknown. However, to recognize the region of the hoof in which the cause of the lameness is located, the Merck Veterinary Manual [14] suggested the following: “when there is pain in the toe, the retraction phase (when hoof passes behind the phase of vertical weight-bearing) of the stride is reduced considerably. In contrast, if the pain is located in the heel, the protraction phase of the stride is reduced or the hoof is not carried as far forward as is normal”. To define whether the cause of lameness is in the hoof or the upper limb, Jackson and Cockcroft [15] stated that supporting limb lameness (i.e., shortened weight bearing and quick swing phase in order to minimize the contact with the ground) is associated with lesions in the hoof whereas swinging leg lameness (limb is held in extension during wing phase as flexion causes pain) with disorders in the upper limb. Yet, it should be noted that abnormal gait may develop not only as a result of disorders in the locomotor system but also disorders in other organs (e.g., severe udder distension in heifers) [16]. Furthermore, both environmental and cow factors may impair locomotion. Wet flooring, dark environments, time relative to claw trimming sessions, age, cow dimensions, lactation and gestation stage might cause normal changes in cow gait that are not related to lameness [17]. In this review, lameness is considered abnormal locomotion due to pain caused by the locomotor system. Although changes in the general behavior of cows, like lying, standing or feeding behavior, Animals 2015,5841 have been associated with lameness, changes in locomotion are the most commonly used and most direct ways to monitor lameness. 3. Normal Locomotion of Dairy Cows To understand abnormal locomotion (i.e., gait and posture), thorough insight into normal gait—with an emphasis on the way cows walk—is essential. Fundamental gait analyses have focused on the relation between limb/hoof events, which are traditionally described in terms of footfall patterns or phase relationships between limbs [18]. Phillips [19] defined walking as strides where each limb is lifted by shortening of the limb through flexion of the joints using especially hip, knee, hock and digital flexor muscles. The limb then enters the swing phase (during which the limb has no contact with the ground) and is placed on the ground through slow extension of the joint. Once the limb is on the ground, it checks the support and the sole is pushed hard against the floor by contracting the digital flexors. This enables the start for the support phase (during which the limb is in contact with the ground) again followed by the next swing phase. The walk is a gait of four evenly spaced beats with no suspension phase (i.e., the phase in which the animal moves forward without any limb touching the ground) but with alternate support by two or three limbs. The usual succession of ground-contacting limbs during walk is LH–LF–RH–RF (where L, R, H, F indicate left, right, hind, and front, respectively) with a regular rhythm and even spacing between footfalls. Hildebrand [20] stated that it is conventional to enter the cycle with the footfall of the left hind foot. To allow the hind limb to overlap the front limb imprint, the latter is usually lifted just before the hind limb is placed. During a normal walk, the duration of support exceeds that of swing, and different limbs show less than 50% overlap in the swing phase of two successive steps [18]. Several techniques have been proposed to schematically represent animal gait, resulting in different ways to describe gait. Muybridge [21] proposed the footfall sequence formula (Figure 1A), while McGhee [22] defined the equivalent gait matrix (Figure 1B). Hildebrand’s [20] representation of the relative duration of the stance and swing phases during one stride are illustrated in Figure 1C. Abourachid [23] introduced an antero-posterior sequence approach in analysing gaits of quadrupeds where—besides focusing on strides—the time lags between the movements of both front limbs (F-lag) or hind limbs (H-lag) or between the front and hind limbs of the same side (P-lag) were added (Figure 1D). If the percentage of time lag during the whole cycle is 50% for both F-lag and H-lag and > 50% for P-lag, the walk is called lateral. Telezhenko [24] proposed track-way diagrams to describe the spatial relation between different limbs (Figure 1E). Detailed information on footfall patterns in Holstein dairy cows was provided by Flower et al. [25] and Telezhenko [24] using a temporal gait diagram and a track-way representation of the spatial distribution of all limbs, respectively. Flower et al. [25] found that for non-lame cows only 18% of the stride time was in triple support, i.e., time when the body weight is supported by three limbs instead of two (Figure 2A). Telezhenko [24] reported three types of possible track-way measurements that either describe the rate of progression (stride length and tracking), the balance (step width, step angle) or the coordination between different limbs (step length, step asymmetry). Some of these variables are visualized in Figure 2B. Animals 2015,5842 Animals 2015, 5 842 A B C D E Figure 1. Schematic representation of quadruped gait: (A) footfall sequence formula according to Muybridge [21] with compact notation according to Hildebrand [20]; (B) gait matrix [22]; (C) gait diagram of a common lateral walk with additional temporal information [20]; (D) track-way diagrams revealing the use of the F-, Hand P-lag of Abourachid [23]; (E) spatial distribution of the supporting limbs [24]. Figure 1. Schematic representation of quadruped gait: (A) footfall sequence formula according to Muybridge [21] with compact notation according to Hildebrand [20]; (B) gait matrix [22]; (C) gait diagram of a common lateral walk with additional temporal information [20]; (D) track-way diagrams revealing the use of the F-, Hand P-lag of Abourachid [23]; (E) spatial distribution of the supporting limbs [24]. Animals 2015,5843 Animals 2015, 5 843 (A) (B) Figure 2. Temporal (A) and spatial (B) representation of cow gait: (A) the mean proportion of double and triple support time during a single stride for each limb; RF = right front hoof, LF = left front hoof, LR = left rear hoof, RR = right rear hoof, hoof strike (HS), and hoofoff (HO) [25]; (B) spatial track way measurements as performed by Telezhenko [24]; FL = front left; RL = rear left; FR = front right, and RR = rear right. Besides spatio-temporal characteristics of gait, the forces exerted by the claws on the ground have been measured. One of the first pressure measurements in cattle was performed by Distl and Mair [26]. They reported an average force of 19 N/cm² with peaks of 56 and 59 N/cm² for first and second parity cows, respectively. Rajkondawar et al. [27] measured peak and averaged ground reaction forces (relative to the cow’s body weight) resulting in dimensionless numbers of 0.41 N/N and 0.30 N/N, respectively, and a stance time of 1.29 s for the hind limbs during walk. The majority of the total body weight (54 to 60%) is supported by the front limbs [19,28,29]. Table 1. Average (±stdev) values of gait variables for non-lame cows. Gait Variable Average ± stdev Walking speed (m/s) 1.350 ± 0.150 1 Stride length (m) 1.591 ± 0.005 2 Stride time (s) 1.523 ± 0.009 2 Stance time (s) 1.011 ± 0.007 2 Step overlap (m) 0.011 ± 0.003 2 Abduction (m) 0.029 ± 0.001 2 Asymmetry in step width (m) 0.201 ± 0.003 2 Asymmetry in step length (m) 0.419 ± 0.002 2 Asymmetry in step time (s) 0.389 ± 0.002 2 Asymmetry in stance time (s) 0.024 ± 0.001 2 Leg weight ratio (/) * 80.0 3 1 Thorup et al. [30], n = 348 cows; 2 Van Nuffel [17] n = 34 cows; 3 Pastell and Kujala [31], n = 73 cows; * Leg weight ratio is the ratio between the heavier and lighter hind leg. Figure 2. Temporal (A) and spatial (B) representation of cow gait: (A) the mean proportion of double and triple support time during a single stride for each limb; RF = right front hoof, LF = left front hoof, LR = left rear hoof, RR = right rear hoof, hoof strike (HS), and hoof-off (HO) [25]; (B) spatial track way measurements as performed by Telezhenko [24]; FL = front left; RL = rear left; FR = front right, and RR = rear right. Besides spatio-temporal characteristics of gait, the forces exerted by the claws on the ground have been measured. One of the first pressure measurements in cattle was performed by Distl and Mair [26]. They reported an average force of 19 N/cm2with peaks of 56 and 59 N/cm2for first and second parity cows, respectively. Rajkondawar et al. [27] measured peak and averaged ground reaction forces (relative to the cow’s body weight) resulting in dimensionless numbers of 0.41 N/N and 0.30 N/N, respectively, and a stance time of 1.29 s for the hind limbs during walk. The majority of the total body weight (54% to 60%) is supported by the front limbs [19,28,29]. Table 1. Average (˘stdev) values of gait variables for non-lame cows. Gait Variable Average ˘stdev Walking speed (m/s) 1.350 ˘0.150 1 Stride length (m) 1.591 ˘0.005 2 Stride time (s) 1.523 ˘0.009 2 Stance time (s) 1.011 ˘0.007 2 Step overlap (m) 0.011 ˘0.003 2 Abduction (m) 0.029 ˘0.001 2 Asymmetry in step width (m) 0.201 ˘0.003 2 Asymmetry in step length (m) 0.419 ˘0.002 2 Asymmetry in step time (s) 0.389 ˘0.002 2 Asymmetry in stance time (s) 0.024 ˘0.001 2 Leg weight ratio (/) * 80.0 3 1Thorup et al. [30], n = 348 cows; 2Van Nuffel [17] n = 34 cows; 3Pastell and Kujala [31], n = 73 cows; * Leg weight ratio is the ratio between the heavier and lighter hind leg. Animals 2015,5844 Mean values of locomotion characteristics of non-lame cows are provided in Table 1. Differences in gait variable measurements reported in the literature might be explained by differences in the breed, cow dimensions and cow factors (parity, locomotion scores, lactation stage, etc.) of the animals in the study, the measurement techniques used, measurement set-up or the definitions of the variable. Telezhenko [24] indeed found significant correlations (R2of 0.51–0.66) between the height and the length of the cow and stride and step length, step angle and width. In a pilot study by Van Nuffel [17], step length and speed for example decreased with age and gestation stage. However, no information on cow dimensions (height, length, breast width, udder size, etc.) was collected during this study. The weight of the cow is carried by the claw. The claw wall bears most of the load during walking, while during standing, most pressure was found at the heel area of the claw [32–34]. In a more detailed study done by van der Tol et al. [35], up to 51% (front limbs) and 37% (hind limbs) of the total body weight of the cows is almost completely borne by the lateral part of the claw compared to the medial part during walk, especially at heel strike. When standing, the posterior part and anterior of the claw are most often subjected to the highest pressure in the front limbs and hind limbs, [36]. Hence, depending on the lesion type, severity and location, a direct association can or cannot be found between the weight applied to the limb and the lesions present [37,38]. Moreover, when cows are lame on both hind limbs, weight is seldom transferred to the front limbs in an attempt to reduce pressure on the painful limbs. In contrast, cows that were lame on both front limbs were found to transfer some of the weight to the hind limbs [28]). However, the fact that painful lesions often occur symmetrically makes detection of lameness even more difficult. Body movement patterns of the head, spine, joints and tail can also be considered characteristics of locomotion [20], yet recent studies of locomotion have mainly focused on the timing and duration of footfalls on the ground and the different patterns of support and swing phases (i.e., gait) and less on posture. 4. Signs of Lameness 4.1. Changes in Gait Patterns In an attempt to detect abnormal or lame gait, several gait characteristics are used in the locomotion scoring of cattle. However, very few papers describe the relationship between poor locomotion and the gait characteristics. Flower et al. [16,25] found that lame cows walked slower, had longer stride durations, shorter strides and a more uneven weight distribution over the limbs than non-lame cows. The triple support time in the gait cycle even doubled for lame cows compared with healthy ones. Comparably, Telezhenko et al. [39] reported that severely lame cows walked more slowly with shorter stride and step length and a smaller step angle (see Figure 2B). Maertens et al. [40] and Blackie et al. [41] also found shorter stride lengths and negative tracking distances in lame cows compared with non-lame cows. Tracking distance (also reported as tracking, tracking up or step overlap) can be defined as overor under-extension of the stride of a hind limb resulting in the hind claw not being placed on the same location as the front claw after initiation of the stride [40,42]. Similarly, increased abduction, meaning the sideways distance between the front foot imprint and the next placement of the hind foot on the same side, has been suggested to be associated with lameness [24,40]. According to Maertens Animals 2015,5845 et al. [40] lame cows showed more asymmetry between left and right limbs in step width, step length, step time, stance time and relative force compared with non-lame cows. Studies measuring stride heights or stiff movement of the joints have been scarce. Nevertheless, such variables of general flexion of the joints have also been suggested to be related to lameness [25,43]. Van Nuffel et al. [44,45] introduced the inconsistency of gait as a possible indicator for lameness based on the hypothesis that cows first occasionally show, e.g., short strides interchanged with ‘normal’ stride lengths before taking shorter strides in general when the lameness becomes more severe. 4.2. Changes in Posture or Body Movement Patterns According to Sprecher et al. [46] and Flower et al. [16,25], a more pronounced arched-back posture is associated with lameness in cattle, both while standing and walking. Head movements or head ‘bobs’ (i.e., nodding, vertical movements of the head as the lame limb makes contact with the ground) have also been mentioned as a lameness characteristic in cattle [16,25,47]. 4.3. Changes in Weight Distribution Patterns Lame animals tend to shift their body weight onto non-affected limbs to reduce pain [48]. Indeed, when ground reaction forces were measured for cows during walking, the average and peak ground force reactions were found to decrease with an increase in locomotion score. This suggests that these forces could be used to discriminate lame from non-lame cows [27,49,50]. In practice, measuring weight distributions between the limbs of cows is more feasible during stance. Not surprisingly, lame cows placed more weight on the healthy limb in comparison to healthy animals which distributed their body weight more evenly [31]. According to Rushen et al. [48], lame cows had more than the normal amount of weight on the limb that was contralateral to the injured limb. The more severe the degree of lameness, the clearer the relationship with the body weight distribution was. However, painful lesions often occur symmetrically, making detection of lameness even more difficult. When cows are lame on both hind limbs, weight is seldom transferred to the front limbs in an attempt to reduce pressure on the painful limbs [28]. In contrast, cows that were lame on both front limbs were found to transfer some of the weight to the hind limbs [28]. These results were confirmed by measuring weight shifting between the contralateral limbs [28,37,38,51,52]. Higher leg weight ratios and higher variations in weight shifts between limbs were successfully used to distinguish lame from non-lame cows [31]. Increased step and kick behavior during milking has also been highlighted as a good indicator for lameness [51,53,54]. Chapinal and Tucker [55] confirmed these findings. In their study, lame cows stepped more with the hind limbs compared with non-lame cows, although this was not the case for the front limbs. The number of steps was found to be closely correlated to the weight shifting between limbs during measurements and with the limb weight ratio. However, in the study by Rousing et al. [56] a higher kicking frequency during milking was associated with pain and discomfort related to teat injuries, whereas no relation between lameness and kicking or stepping behavior was found. Animals 2015,5846 4.4. Changes in Behavior Besides locomotion, the most important natural behaviors for cow health, welfare and productivity are resting, eating, ruminating, and socializing [57]. Lameness has been associated with longer lying times, longer periods of standing in alleys and decreased feeding behavior [58–60] As lame cows tend to have pain and be lower in rank [61], this results in different behaviors such as longer duration spend at the resting, shorter duration at feeding places or grazing [53,62]. Detection of abnormal behavior is out of the scope of this review, but a detailed overview of the link between lameness and changes in lying/standing behavior, activity, feeding behavior, behavior around milking, estrus and social behavior can be found in Van Nuffel [17]. 4.5. Some Considerations Several gait and behavioral characteristics change when cows develop lameness making them interesting characteristics to be used by automatic lameness detection systems. To implement them, automating the assessment of these characteristics is a prerequisite. Direct characteristics of gait (e.g., stride length) might show more sensitivity in detecting lameness compared with more indirect characteristics such as lying and feeding behavior. Indeed, the intrinsic problem with using indirect measures of lameness is that they are also influenced by other factors, thus inhibiting straightforward conclusions of causal effects. Lying behavior for example is also associated with stage of lactation, production level, lying surface, bedding, cow comfort, stocking density, stall size, stall configuration, pen lay out, milking and feeding management, social rank, etc. [63]. A number of these factors might also influence the ‘normal’ gait of cows as recently investigated by Van Nuffel [17]. Also, it is not clear what the cause and effect relationship is between lameness and behavior; lameness may alter behavior, or may be the consequence of the behavior [64]? 5. Visual Locomotion Scoring To reduce lameness, farmers need to be aware of the number of lame cows and the severity of lameness in their herd. The commonly accepted methodologies to quantify lameness rely on spotting changes in gait, posture or behavior of the cows. In practice, this is done using subjective methods such as visual observations leading to locomotion scorings by the farmer, an employee, a veterinarian or an agricultural consultant. Subjective scoring is quick to apply, inexpensive and easy-to-use. As many as 25 different visual scoring systems are available. Most of them are based on assessing walking cows, but they differ with respect to the scale used, the gait characteristics and postures considered and the definition of lameness applied [65]. The used scales range from binary (lame versus non-lame) to continuous (1 to 100). On a 5-point scale from 1 to 5, cows with a score > 3 are typically defined as being clinically lame. A range of indicators, defined as specific characteristics of gait or posture relevant for lameness, are used to help the observer assess the quality of a cow’s gait and posture and score their gait in classes ranging from non-lame to severely lame. Animals 2015,5853 assessing lameness of individual cows in a herd can be practically challenging. Although probably possible to construct, there is generally no specially designed alley available on farms where cows pass daily while their gait is observed. Cows often have to be assessed in barns lacking sufficient space to walk in a straight line for several meters. The slatted floor is furthermore mostly covered with slippery manure impairing normal locomotion. Also, cows can be reluctant to walk past herd mates and need to be motivated to walk past them. This again disturbs their normal locomotion and increases the need for an additional helper to guide the cow in order for the observer to give a score. Being a prey species and hence stoic in nature, lame cows might also mask signs of vulnerability, because this would make them an easier target for predators [94]. Therefore, in the presence of an observer or when forced to walk, cows tend to only show lame locomotion and behavior when the lameness is at an advanced stage. As only a trained observer may notice the onset of lameness by multiple subtle gait aberrations, detecting early lameness tends to be difficult. In addition, as discussed above, practical guidelines on the set-up for visual locomotion scoring are lacking. Most visual scoring systems are not designed to assess lameness on a day-to-day basis, but to compare locomotion scores between different treatments or set-ups. Sporadic scoring of the entire herd by an experienced observer only provides the farmer or veterinarian with prevalence information for that specific moment. Because of the non-straightforward link between lameness and claw lesions or other diseases of the bones and joints [95], the usefulness of locomotion scores is limited to indications of pain, rather than as a diagnostic tool for the presence of hoof lesions [68]. This makes a thorough check of the cow necessary. More importantly, all cows have their own specific way of walking [17] and reacting on illness [96]. Hence, a detection system should rather be based on changes in indicators over time than on the deviation from the group mean. Daily monitoring of locomotion or behavior on farm by a trained observer is too time-consuming and thus very cost inefficient. Automatic measurement of lameness-related, animal-based characteristics would allow for daily measurements and could, therefore, be a better option. 6. Conclusions Clear and accurate definitions including numerical information of normal and abnormal cow gait are lacking in the description of visual locomotion scoring system. This adds to the difficulty of performing visual locomotion scoring to assess lameness on farms. Moreover, the farm design, experimental set-ups and presence of an observer impede proper scoring on practical farms even more. Different practical approaches are used during locomotion scoring, resulting in incomparable results between studies. The different set-ups used and their influence on the resulting locomotion score was evaluated. Automated measurements of lameness-related cow features might therefore solve some of these issues. However, the practical set-up of such systems will also be challenging. Author Contributions This review was written by the first author; all co-authors made suggestions and alterations to the review for improvement. Animals 2015,5854 Conflicts of Interest The authors declare no conflict of interest. References 1. International Dairy Federation (IDF). World Dairy Situation 2013; IDF: Brussels, Belgium, 2013. 2. Van der Straeten, B.; Deuninck, J.; Van Gijseghem, D. De melkproductie in Vlaanderen na 2015; beleidsdomein Landbouw en Visserij, afdeling Monitoring en Studie: Brussel, Belgium, 2012. 3. Enting, H.; Kooij, D.; Dijkhuizen, A.A.; Huirne, R.B.M.; Noordhuizen-Stassen, E.N. Economic losses due to clinical lameness in dairy cattle. Livest. Prod. Sci. 1997,49, 259–267. [CrossRef] 4. Laven, R.A. Lameness on New Zealand dairy farms: Perception and reality. In Proceedings of the 17th International Symposium and 9th International Conference on Lameness in Ruminants, Bristol, UK, 11–14 August 2013; pp. 25–27. 5. Gröhn, Y.T.; Rajala-Schultz, P.J.; Allore, H.G.; DeLorenzo, M.A.; Hertl, J.A.; Galligan, D.T. Optimizing replacement of dairy cows: Modeling the effects of diseases. Prev. Vet. Med. 2003,61, 27–43. [CrossRef] 6. English, A.M.; Ullrich, E.; Bergfeld, U.; Fleisher, J.; Roesler, U.; Mueller, K.E. Animal suitability index and lameness in dairy farms in Saxony, Germany. In Proceedings of the 17th International Symposium and 9th International Conference on Lameness in Ruminants, Bristol, UK, 11–14 August 2013; pp. 109–110. 7. Huxley, J.N. Impact of lameness and claw lesions in cows on health and production. Livest. Sci. 2013,156, 64–70. [CrossRef] 8. Leach, K.A.; Paul, E.S.; Whay, H.R.; Barker, Z.E.; Maggs, C.M.; Sedgwick, A.K.; Main, D.C.J. Reducing lameness in dairy herds—Overcoming some barriers. Res. Vet. Sci. 2013,94, 820–825. [CrossRef] [PubMed] 9. Van Nuffel, A.; Zwertvaegher, I.; Van Weyenberg, S.; Pastell, M.; Thorup, V.M.; Bahr, C.; Sonck, B.; Saeys, W. Lameness detection in dairy cows: Part 2. Use of sensors to automatically register changes in locomotion or behavior. Animals 2015. [CrossRef] 10. Straw, B.E.; Dewe, C.E.; Wilson, M.R. Differential diagnosis of swine diseases. In Diseases of Swine; Straw, B.E., D’Allaire, S., Mengeling, W.L., Taylor, D.J., Eds.; Iowa State University Press: Ames, IA, USA, 1999; pp. 41–86. 11. Maas, J. Musculoskeletal abnormalities. In Large Animal Internal Medicine; Smith, B.P., Ed.; Mosby, Inc.: ST. Louis, MS, USA, 2009; pp. 217–231. 12. Beusker, N. Welfare of Dairy Cows: Lameness in Cattle. Literature Review; The Institute of Animal Hygiene, Animal Welfare and Animal Ethology, School of Veterinary Medicine: Hannover, Germany, 2007. 13. Greenough, P.R. Bovine Laminitis and Lameness, a Hands-on Approach; Bergsten, C., Brizzi, A., Mülling, C., Eds.; Saunders Elsevier: Amsterdam, The Netherlands, 2007. 14. Merck, 2015. The Merck Veterinary Manual. Lameness in Cattle. Available online: http://www. merckvetmanual.com/mvm/musculoskeletal_system/lameness_in_cattle/overview_of_lameness_in_ cattle.html (accessed on 13 July 2015). Animals 2015,5855 15. Jackson, P.G.G.; Cockcroft, P.D. Clinical examination of the musculoskeletal system. In Clinical Examination of Farm Animals; Blackwell Publishing: Oxford, UK, 2002; pp. 167–197. 16. Flower, F.C.; Sanderson, D.J.; Weary, D.M. Effects of milking on dairy cow gait. J. Dairy Sci. 2006,89, 2084–2089. [CrossRef] 17. Van Nuffel, A. Development of an Automated Detection System for Lameness in Cattle: The Gaitwise System. Ph.D. Thesis, Katholic University of Leuven, Leuven, Belgium, 2014. 18. Eilam, D. Postnatal development of body architecture and gait in several rodent species. J. Exp. Biol. 1997,200, 1339–1350. [PubMed] 19. Hildebrand, M. The quadrupedal gaits of vertebrates. Bioscience 1989,39, 766–775. [CrossRef] 20. Muybridge, E. Animals in Motion; Dover Publications: New York, NY, USA, 1899. 21. McGhee, R.B. On the stability of quadruped gaits. Math. Biosci. 1968,3, 331–351. [CrossRef] 22. Abourachid, A. A new way of analysing symmetrical and asymmetrical gaits in quadrupeds. Comptes Rendus Biol. 2003,326, 625–630. [CrossRef] 23. Telezhenko, E. Measurement of spatial gait parameters from footprints of dairy cows. Animal 2009, 3, 1746–1753. [CrossRef] [PubMed] 24. Flower, F.C.; Sanderson, D.J.; Weary, D.M. Hoof pathologies influence kinematic measures of dairy cow gait. J. Dairy Sci. 2005,88, 3166–3173. [CrossRef] 25. Distl, O.; Mair, A. Computerized analysis of pedobarometric forces in cattle at the ground surface/floor interface. Comp. Electr. Agric. 1993,8, 237–250. [CrossRef] 26. Rajkondawar, P.G.; Liu, M.; Dyer, R.M.; Neerchal, N.K.; Tasch, U.; Lefcourt, A.M.; Erez, B.; Varner, M.A. Comparison of models to identify lame cows based on gait and lesion scores, and limb movement variables. J. Dairy Sci. 2006,89, 4267–4275. [CrossRef] 27. Neveux, S.; Weary, D.M.; Rushen, J.; von Keyserlingk, M.A.G.; de Passille, A.M. Hoof discomfort changes how dairy cattle distribute their body weight. J. Dairy Sci. 2006,89, 2503–2509. [CrossRef] 28. Pastell, M.; Hautala, M.; Poikalainen, V.; Praks, J.; Veermae, I.; Kujala, M.; Ahokas, J. Automatic observation of cow leg health using load sensors. Comp. Electr. Agr. 2008,62, 48–53. [CrossRef] 29. Thorup, V.M.; Munksgaard, L.; Robert, P.-E.; Erhard, H.W.; Thomsen, P.T.; Friggens, N.C. Lameness detection via leg-mounted accelerometers on dairy cows on four commercial farms. Animal 2015. [CrossRef] [PubMed] 30. Pastell, M.E.; Kujala, M. A probabilistic neural network model for lameness detection. J. Dairy Sci. 2007,90, 2283–2292. [CrossRef] [PubMed] 31. Phillips, C. Locomotion and movement. In Cattle Behaviour & Welfare; Phillips, C., Ed.; Blackwell Science Ltd: Malden, MA, USA, 2007; Chapter 12. pp. 179–197. 32. Mair, A.; Diebschlag, W.; Distl, O.; Krausslich, H. Measuring device for the analysis of pressure distribution on the foot soles of cattle. Journal of Veterinary Medicine Series A-Zentralblatt fur Veterinarmedizin Reihe A-Physiology Pathology Clinical Medicine 1988,35, 696–704. 33. Scott, G.B. Lameness and pregnancy in friesian dairy-cows. Brit. Vet. J. 1988,144, 273–281. [CrossRef] Animals 2015,5856 34. Distl, O.; Koorn, D.S.; McDaniel, B.T.; Peterse, D.; Politiek, R.D.; Reurink, A. Claw traits in cattle breeding programs: Report of the E.A.A.P. Working Group “Claw Quality in Cattle”. Livest. Prod. Sci. 1990,25, 1–13. [CrossRef] 35. Van der Tol, P.P.J.; Metz, J.H.M.; Noordhuizen-Stassen, E.N.; Back, W.; Braam, C.R.; Weijs, W.A. The vertical ground reaction force and the pressure distribution on the claws of dairy cows while walking on a flat substrate. J. Dairy Sci. 2003,86, 2875–2883. [CrossRef] 36. Van der Tol, P.P.J.; Metz, J.H.M.; Noordhuizen-Stassen, E.N.; Back, W.; Braam, C.R.; Weijs, W.A. The pressure distribution under the bovine claw during square standing on a flat substrate. J. Dairy Sci. 2002,85, 1476–1481. [CrossRef] 37. Neveux, S.; Oostra, J.; de Passille, A.M.; Rushen, J. Validating on-farm tools for their ability to detect lameness in dairy cows. In Proceedings of the 37th International congress of the ISAE, Bano Therme, Italy, 24–28 June 2003. 38. Pastell, M.; Hanninen, L.; de Passille, A.M.; Rushen, J. Measures of weight distribution of dairy cows to detect lameness and the presence of hoof lesions. J. Dairy Sci. 2010,93, 954–960. [CrossRef] [PubMed] 39. Telezhenko, E.; Bergsten, C. Influence of floor type on the locomotion of dairy cows. Appl. Animal Behav. Sci. 2005,93, 183–197. [CrossRef] 40. Maertens, W.; Vangeyte, J.; Baert, J.; Jantuan, A.; Mertens, K.C.; De Campeneere, S.; Pluk, A.; Opsomer, G.; Van Weyenberg, S.; Van Nuffel, A. Development of a real time cow gait tracking and analysing tool to assess lameness using a pressure sensitive walkway: The GAITWISE system. Biosyst. Eng. 2011,110, 29–39. [CrossRef] 41. Blackie, N.; Bleach, E.C.L.; Amory, J.R.; Scaife, J.R. Associations between locomotion score and kinematic measures in dairy cows with varying hoof lesion types. J. Dairy Sci. 2013,96, 3564–3572. [CrossRef] [PubMed] 42. Shearer, J.K.; Stock, M.L.; van Amstel, S.R.; Coetzee, J.F. Assessment and management of pain associated with lameness in cattle. Vet. Clin. North Am. Food Animal Pract. 2013,29, 135–165. [CrossRef] [PubMed] 43. Pluk, A.; Bahr, C.; Poursaberi, A.; Maertens, W.; Van Nuffel, A.; Berckmans, D. Automatic measurement of touch and release angles of the fetlock joint for lameness detection in dairy cattle using vision techniques. J. Dairy Sci. 2012,95, 1738–1748. [CrossRef] [PubMed] 44. Van Nuffel, A.; Vangeyte, J.; Mertens, K.C.; Pluym, L.; De Campeneere, S.; Saeys, W.; Opsomer, G.; Van Weyenberg, S. Exploration of measurement variation of gait variables for early lameness detection in cattle using the GAITWISE. Livest. Sci. 2013,156, 88–95. [CrossRef] 45. Van Nuffel, A.; Saeys, W.; Sonck, B.; Vangeyte, J.; Mertens, K.C.; De Ketelaere, B.; Van Weyenberg, S. Variables of gait inconsistency outperform basic gait variables in detecting mildly lame cows. Livest. Sci. 2015,177, 125–131. [CrossRef] 46. Sprecher, D.J.; Hostetler, D.E.; Kaneene, J.B. A lameness scoring system that uses posture and gait to predict dairy cattle reproductive performance. Theriogenology 1997,47, 1179–1187. [CrossRef] 47. Nordlund, K.V.; Cook, N.B.; Oetzel, G.R. Investigating strategies for laminitis in problem herds. J. Dairy Sci. 2004,87, E27–E35. [CrossRef] Animals 2015,5857 48. Rushen, J.; Pombourcq, E.; de Passille, A.M. Validation of two measures of lameness in dairy cows. Appl. Animal Behav. Sci. 2007,106, 173–177. [CrossRef] 49. Scott, G.B. Changes in limb loading with lameness for a number of friesian cattle. Brit. Vet. J. 1989,145, 28–38. [CrossRef] 50. Rajkondawar, U.; Lefcourt, A.M.; Neerchal, N.K.; Dyer, R.M.; Varner Erez, B.; Tasch, P.G. The development of an objective lameness scoring system for dairy herds: Pilot study. Trans. ASABE 2002,45, 1123–1125. [CrossRef] 51. Chapinal, N.; de Passille, A.M.; Rushen, J.; Wagner, S. Effect of analgesia during hoof trimming on gait, weight distribution, and activity of dairy cattle. J. Dairy Sci. 2010,93, 3039–3046. [CrossRef] [PubMed] 52. Chapinal, N.; de Passille, A.M.; Rushen, J.; Wagner, S. Automated methods for detecting lameness and measuring analgesia in dairy cattle. J. Dairy Sci. 2010,93, 2007–2013. [CrossRef] [PubMed] 53. Hassall, S.A.; Ward, W.R.; Murray, R.D. Effects of lameness on the behavior of cows during the summer. Vet. Rec. 1993,132, 578–580. [CrossRef] [PubMed] 54. Pastell, M.; Aisla, A.M.; Hautala, M.; Poikalainen, V.; Praks, J.; Veermae, I.; Ahokas, J. Contactless measurement of cow behavior in a milking robot. Behav. Res. Meth. 2006,38, 479–483. [CrossRef] 55. Chapinal, N.; Tucker, C.B. Validation of an automated method to count steps while cows stand on a weighing platform and its application as a measure to detect lameness. J. Dairy Sci. 2012,95, 6523–6528. [CrossRef] [PubMed] 56. Rousing, T.; Bonde, M.; Badsberg, J.H.; Sorensen, J.T. Stepping and kicking behaviour during milking in relation to response in human-animal interaction test and clinical health in loose housed dairy cows. Livest. Prod. Sci. 2004,88, 1–8. [CrossRef] 57. Krawczel, P.; Grant, R. Effectss of cow comfort on milk quality, productivity and behavior. In Proceedings of the NMC Annual Meeting, Charlotte, NC, USA, 25–28 January 2009; pp. 15–24. 58. Singh, S.S.; Ward, W.R.; Lautenlach, L.; Hughes, J.W.; Murray, R.D. Behaviour of first lactation and adult dairy cows while housed and at pasture and its relationship with sole lesions. Vet. Rec. 1993,133, 469–474. [CrossRef] [PubMed] 59. Gonzalez, L.A.; Tolkamp, B.J.; Coffey, M.P.; Ferret, A.; Kyriazakis, I. Changes in feeding behavior as possible indicators for the automatic monitoring of health disorders in dairy cows. J. Dairy Sci. 2008,91, 1017–1028. [CrossRef] [PubMed] 60. Blackie, N.; Bleach, E.; Amory, J.; Scaife, J. Impact of lameness on gait characteristics and lying behaviour of zero grazed dairy cattle in early lactation. Appl. Animal Behav. Sci. 2011,129, 67–73. [CrossRef] 61. Galindo, F.; Broom, D.M. The relationships between social behaviour of dairy cows and the occurrence of lameness in three herds. Res. Vet. Sci. 2000,69, 75–79. [CrossRef] [PubMed] 62. Yunta, C.; Guasch, I.; Bach, A. Short communication: Lying behavior of lactating dairy cows is influenced by lameness especially around feeding time. J. Dairy Sci. 2012,95, 6546–6549. [CrossRef] [PubMed] 63. Mattachini, G.; Riva, E.; Provolo, G. The lying and standing activity indices of dairy cows in free-stall housing. Appl. Animal Behav. Sci. 2011,129, 18–27. [CrossRef] Animals 2015,5858 64. Cook, N.B.; Nordlund, K.V. The influence of the environment on dairy cow behavior, claw health and herd lameness dynamics. Vet. J. 2009,179, 360–369. [CrossRef] [PubMed] 65. Schlageter-Tello, A.; Bokkers, E.A.; Groot Koerkamp, P.W.; Van Hertem, T.; Viazzi, S.; Romanini, C.E.; Halachmi, I.; Bahr, C.; Berckmans, D.; Lokhorst, K. Effect of merging levels of locomotion scores for dairy cows on intraand interrater reliability and agreement. J. Dairy Sci. 2014,97, 5533–5542. [CrossRef] [PubMed] 66. Manson, F.J.; Leaver, J.D. The influence of concentrate amount on locomotion and clinical lameness in dairy cattle. Animal Prod. 1988,47, 185–190. [CrossRef] 67. Wells, S.J.; Trent, A.M.; Marsh, W.E.; Robinson, R.A. Prevalence and severity of lameness in lactating dairy-cows in a sample of Minnesota and Wisconsin Herds. J. Am. Vet. Med. Assoc. 1993,202, 78–82. [PubMed] 68. Winckler, C.; Willen, S. The reliability and repeatability of a lameness scoring system for use as an indicator of welfare in dairy cattle. Acta Agr. Scand. A Animal Sci. 2000,30, 103–107. [CrossRef] 69. Breuer, K.; Hemsworth, P.H.; Barnett, J.L.; Matthews, L.R.; Coleman, G.J. Behavioural response to humans and the productivity of commercial dairy cows. Appl. Animal Behav. Sci. 2000,66, 273–288. [CrossRef] 70. Flower, F.C.; Weary, D.M. Effect of hoof pathologies on subjective assessments of dairy cow gait. J. Dairy Sci. 2006,89, 139–146. [CrossRef] 71. Dyer, R.M.; Neerchal, N.K.; Tasch, U.; Wu, Y.; Dyer, P.; Rajkondawar, P.G. Objective determination of claw pain and its relationship to limb locomotion score in dairy cattle. J. Dairy Sci. 2007,90, 4592–4602. [CrossRef] [PubMed] 72. Welfare Qualityr.Welfare QualityrAssessment Protocol for Cattle; Welfare Qualityr Consortium: Lelystad, The Netherlands, 2009. 73. Van Nuffel, A.; Sprenger, M.; Tuyttens, F.; Maertens, W. Can people see gait irregularities? Animal Welf. 2009,19, 433–439. 74. Leach, K.A.; Dippel, S.; Huber, J.; March, S.; Winckler, C.; Whay, H.R. Assessing lameness in cows kept in tie-stalls. J. Dairy Sci. 2009,92, 1567–1574. [CrossRef] [PubMed] 75. Gleeson, D.E.; O’Brien, B.; Boyle, L.; Earley, B. Effect of milking frequency and nutritional level on aspects of the health and welfare of dairy cows. Animal 2007,1, 125–132. [CrossRef] [PubMed] 76. Olmos, G.; Boyle, L.; Horan, B.; Berry, D.P.; O’Connor, P.; Mee, J.F.; Hanlon, A. Effect of genetic group and feed system on locomotion score, clinical lameness and hoof disorders of pasture-based Holstein-Friesian cows. Animal 2009,3, 96–107. [CrossRef] [PubMed] 77. O’Driscoll, K.; Gleeson, D.; O’Brien, B.; Boyle, L. Effect of milking frequency and nutritional level on hoof health, locomotion score and lying behaviour of dairy cows. Livest. Sci. 2010,127, 248–256. [CrossRef] 78. O’Driscoll, K.K.M.; Schutz, M.M.; Lossie, A.C.; Eicher, S.D. The effect of floor surface on dairy cow immune function and locomotion score. J. Dairy Sci. 2009,92, 4249–4261. [CrossRef] [PubMed] 79. O’Callaghan, K.A.; Cripps, P.J.; Downham, D.Y.; Murray, R.D. Subjective and objective assessment of pain and discomfort due to lameness in dairy cattle. Animal Welf. 2003,12, 605–610. Animals 2015,5859 80. Phillips, C.J.C.; Morris, I.D. The locomotion of dairy cows on concrete floors that are dry, wet, or covered with slurry of excreta. J. Dairy Sci. 2000,83, 1767–1772. [CrossRef] 81. Rushen, J.; de Passille, A.M. Effects of roughness and compressibility of flooring on cow locomotion. J. Dairy Sci. 2006,89, 2965–2972. [CrossRef] 82. Chapinal, N.; de Passille, A.M.; Pastell, M.; Hanninen, L.; Munksgaard, L.; Rushen, J. Measurement of acceleration while walking as an automated method for gait assessment in dairy cattle. J. Dairy Sci. 2011,94, 2895–2901. [CrossRef] [PubMed] 83. Van Der Tol, P.P.J.; Metz, J.H.M.; Noordhuizen-Stassen, E.N.; Back, W.; Braam, C.R.; Weijs, W.A. Frictional forces required for unrestrained locomotion in dairy cattle. J. Dairy Sci. 2005,88, 615–624. [CrossRef] 84. Phillips, C.J.C.; Morris, I.D. The locomotion of dairy cows on floor surfaces with different frictional properties. J. Dairy Sci. 2001,84, 623–628. [CrossRef] 85. Hollenbeck, A.R. Problems of reliability in observational research. In Observing Behaviour Vol II—Data Collection and Analysis Methods; Sacket, G.P., Ed.; University Park Press: Baltimore, MD, USA, 1978; pp. 79–98. 86. Brenninkmeyer, C.; Dippel, S.; March, S.; Brinkmann, J.; Winckler, C.; Knierim, U. Reliability of a subjective lameness scoring system for dairy cows. Animal Welf. 2007,16, 127–129. 87. March, S.; Brinkmann, J.; Winkler, C. Effect of training on the inter-observer reliability of lameness scoring in dairy cattle. Animal Welf. 2007,16, 131–133. 88. Main, D.C.J.; Clegg, J.; Spatz, A.; Green, L.E. Repeatability of a lameness scoring system for finishing pigs. Vet. Rec. 2000,147, 574–576. [CrossRef] [PubMed] 89. Channon, A.J.; Walker, A.M.; Pfau, T.; Sheldon, I.M.; Wilson, A.M. Variability of Manson and Leaver locomotion scores assigned to dairy cows by different observers. Vet. Rec. 2009,164, 388–392. [CrossRef] [PubMed] 90. Engel, B.; Bruin, G.; Andre, G.; Buist, W. Assessment of observer performance in a subjective scoring system: Visual classification of the gait of cows. J. Agric. Sci. 2003,140, 317–333. [CrossRef] 91. Thomsen, P.T.; Munksgaard, L.; Tøgersen, F.A. Evaluation of a lameness scoring system for dairy cows. J. Diary Sci. 2008,91, 119–126. [CrossRef] [PubMed] 92. Tuyttens, F.A.M.; Sprenger, M.; Van Nuffel, A.; Maertens, W.; Van Dongen, S. Reliability of categorical versus continuous scoring of welfare indicators: Lameness in cows as a case study. Animal Welf. 2009,18, 399–405. 93. Von Keyserlingk, M.A.G.; Rushen, J.; de Passillé, A.M.; Weary, D.M. Invited review: The welfare of dairy cattle—Key concepts and the role of science. J. Dairy Sci. 2009,92, 4101–4111. [CrossRef] [PubMed] 94. Weary, D.M.; Huzzey, J.M.; von Keyserlingk, M.A.G. BOARD-INVITED REVIEW: Using behavior to predict and identify ill health in animals. J. Animal Sci. 2009,87, 770–777. [CrossRef] [PubMed] 95. Tadich, N.; Flor, E.; Green, L. Associations between hoof lesions and locomotion score in 1098 unsound dairy cows. Vet. J. 2010,184, 60–65. [CrossRef] [PubMed] Animals 2015,5860 96. Kramer, E.; Cavero, D.; Stamer, E.; Krieter, J. Mastitis and lameness detection in dairy cows by application of fuzzy logic. Livest. Sci. 2009,125, 92–96. [CrossRef] 97. Whay, H.R.; Waterman, A.E.; Webster, A.J.F. Associations between locomotion, claw lesions and nociceptive threshold in dairy heifers during the peri-partum period. Vet. J. 1997,154, 155–161. [CrossRef] © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).