Anatomical variations of the deep head of Cruveilhier of the flexor pollicis brevis and its significance for the evolution of the precision grip

Abstract

Cruveilhier described in 1834 the human flexor pollicis brevis (FPB), a muscle of the thenar compartment, as having a superficial and a deep head, respectively, inserted onto the radial and ulnar sesamoids of the thumb. Since then, Cruveilhier’s deep head has been controversially discussed. Often this deep head is confused with Henle’s “interosseous palmaris volaris” or said to be a slip of the oblique adductor pollicis. In the 1960s, Day and Napier described anatomical variations of the insertions of Cruveilhier’s deep head, including its absence, and hypothesized, that the shift of the deep head’s insertion from ulnar to radial facilitated “true opposability” in anthropoids. Their general thesis for muscular arrangements underlying the power and precision grip is sound, but they did not delineate their deep head from Henle’s muscle or the adductor pollicis, and their description of the attachments of Cruveilhier’s deep head were too vague and not supported by a significant portion of the anatomical literature. Here, we reinvestigated Cruveilhier’s deep head to resolve the controversy about it and because many newer anatomy textbooks do not describe this muscle, while it is often an obvious functionally (writing, texting, precision grip) and clinically significant thenar muscle. For the first time, we empirically delineated Cruveilhier’s deep head from neighboring muscles with which it was previously confused. We observed 100% occurrence of the uncontested deep head in 80 human hands, displaying a similar variability of insertions as Day and Napier, but in significantly different numbers. Furthermore, we found variability in the origin and included as important landmarks the trapezoid and the ligamentum carpi radiatum. We tested the assertion regarding the evolutionary morphology and its role in the improvements in thumb movements during various precision grips. Our overall conclusions differ with respect to the developmental and evolutionary origin of the FPB heads.

A smartphone-based online system for fall detection with alert notifications and contextual information of real-life falls

“…5 ). Only three systems [ 30 , 31 , 36 ] resulted in higher sensitivity (80.0-83.3 %) than our system (73.0 %). However, the precision scores in these studies (0.2–22.2 %) were lower than our system (37.5 %).…”

“…This study is the first prospective study that shows the ability to detect real-life falls in real time using a smartphone-based system. Nevertheless, it can be compared to previous prospective studies [ 30 – 36 ] which developed real-life fall systems using dedicated accelerometer sensors (Fig. 5 ).…”

“…Only a small number of prospective studies deployed fall detection systems into the community and validated their performance in real-time [ 29 ]. The majority of systems that tested their fall detection systems in real life have used inertial sensors, commonly accelerometers with or without gyroscopes, for detecting the falls [ 30 – 36 ]. However, all of these systems currently require the use of dedicated sensors which are externally attached to the users’ bodies.…”

“…However, all of these systems currently require the use of dedicated sensors which are externally attached to the users’ bodies. Using these sensors required technical intervention and assistance to attach or remove the sensors [ 30 ], charge or replace the sensor [ 33 ], or download the data [ 31 ], making the process less feasible and non-user-friendly. Additionally, these fall systems are only effective when the device is being utilized and incorporated into the user’s daily life.…”

“…Machine learning algorithms are increasingly used in the development of fall detection models [ 17 ], yet currently all fall-detection machine-learning studies but one [ 30 ] are retrospective. Development and validation of machine learning models in retrospective introduces a risk of overfitting (leakage), where algorithms are effectively using their developers’ knowledge of the test set.…”