**3. Results**

### **3.1 Diagnosis and rehabilitation of sarcopenia**

#### *3.1.1 Assessment of sarcopenia*

#### *3.1.1.1 Accelerometer and actigraph technology in wearable inertial sensors*

Nowadays, wearable inertial sensors have the potential to assess MQ and PP (**Table 1**) [15].

In 2018, the American Academy of Sleep Medicine recommended using the actigraphy test for the diagnosis of sleep disorders [16]. Subsequently, on the basis of the ascertained association between frailty domains and functional limitations [8, 12, 17], Pana et al. investigated the relationship between sleep quality and MS among community-dwelling middle-aged and older adults [12]. The existence of a correlation between sleep disorders and sarcopenia can be expected but, until now, research in this field has been fragmented and no studies have been carried out investigating a possible direct correlation between sleep disorders and sarcopenia. For example, a study [18] has been published on the correlation between peak oxygen consumption and muscle loss. Physiological data were obtained through a feature of the actigraphy test called Actihear [19] which, however, focused on muscular functionality and not on sleep quality.

Accelerometer has been proposed in wearable devices to assess different parameters of physical activity following the "The Physical Activity Guidelines for Americans" (PAG, 2nd edition) [13], as shown in **Table 1**. However, in two studies in which the accelerometry was used, the accelerometry threshold did not prove to be indicative [10, 11]. Viecelli et al. [20] used a smartphone built-in accelerometer to obtain scientific mechano-biological descriptors of resistance exercise training. They aimed to test whether accelerometer data obtained from standard smartphone placed on the weight stack of resistance exercise machines can be used to extract single repetition, contraction-phase specific and total time-under-tension (TUT) [20]. Total time-under-tension is an important mechano-biological descriptor of resistance exercise as it was shown that it is highly positive correlated (R2 = 0.99) with the phosphorylation c-Jun N-terminal kinase (JNK) [21]. Activated JNK phosphorylates the transcription factor SMAD2, leading to the inhibition of myostatin [22], a potent negative regulator of muscle mass [23, 24]. The JNK/SMAD signaling axis is activated by resistance exercise and hence the molecular switch JNK stimulates muscle fiber growth, resulting in increased muscle mass [22] being a direct countermeasure of the muscle loss seen in sarcopenia.

Burd et al. [25] examined the influence of muscle time-under-tension on myofibrillar protein synthesis. Eight young men were allocated into two groups. One group performed three sets of unilateral knee extension at 30% of 1-repetion maximum involving concentric and eccentric muscle actions that were 6 s in duration to failure. The control group performed a work-matched bout that comprised concentric and eccentric actions that were 1 s in duration. As work was matched, the groups significantly differed in time-under-tension (P < 0.001) whereby the slow group had a time-under-tension of


*CSS, cross-sectional studies; CS, cohort studies; SR, systematic review; R-CSS, retrospective-cross-sectional studies; OA; older adults; PA, physical activity; CPM, counts per minute; MVPA, moderate to vigorous PA; HE, health education; DH, digital handgrip; DYN, dynamometer; ACC, accelerometer; ASD, actigraphy sleep diary; WAM, wearable activity monitor; IS, inertial sensors; AMM, anthropometrics muscle mass; S-ACC, strength accelerometer; CF, cognitive function; SPPB, short physical performance battery; TUG, time-up and go test; EFs, executive functions; MS, maximal strength; RFD, rate of force development; SA, strength asymmetry; BS, bilateral strength; FSFT-ST, force steadiness fatigability task-specific tremor; HGS, hand grip strength; GS, gait speed; SAss, sleep assessment; DL, deep learning.*

#### **Table 1.**

*General overview of the paper focused on the accelerometer in wearable devices and the actual use of actigraphy to assess sarcopenia in primary prevention.*

407 ± 23 s and the control group a time-under-tension 50 ± 3, respectively. Myofibrillar protein synthetic rate was higher in the slow condition versus the control condition after 24–30 h recovery (P < 0.001). Therefore, a longer time-under-tension increased myofibrillar protein synthesis longer and to a greater extent than under the control condition.

As evident, time-under-tension is not only an important mechano-biological descriptor of resistance exercise but also of high clinical relevance.

Lastly, a very recent article [14] aimed at identifying and elaborating parameters from gait signals produced by the sensors in order to develop a screening and classification method for sarcopenia. In the study were used specific parameters that they interpreted through an artificial intelligence (AI) model called SHAP (*Shapley Additive Explanations*). The input that applied the SHAP to the descriptive statistical parameters


*RCT, randomized control trials; CSS, cross-sectional studies; ES, exploratory study; VHP, voluntary health people; HY, healthy young; OA, older adults; BIA, biological impedance analysis; HGS, hand grip strength; EMG, electromyography; EEG, electroencephalogram; L&NBRM, logistic & negative binomial regression models; BED, back extension dynamometer; RET, resistance exercise training; KE, knee extension; SCT, stair climbing task; MVC, maximal voluntary contraction; 15mWT, 15 minutes walking task; TXACC-SI, triaxial accelerometer-sensor integrated; QM, quadriceps muscle; BW-RET, lower-limb RET through body-weight squats; MN-RET, seated knee extensions on machine; AAE, adhesive anode electrode; EB-RET, seated knee extensions via elastic bands; S2S, Spike2 Software; FP, frailty phenotype; FI, frailty index; CMAP, compound muscle action potential; MUP, motor unit potential; MNRT, motor unit number-recruitment threshold; MFRRT, motor unit firing rate-recruitment threshold; FRU, firing rate per unit force; MFR, motor unit firing rate; MFD, muscle fiber discrimination; PASE, physical activity of senior elder; IPAQ, International Physical Activity Questionnaire.*

#### **Table 2.**

*General overview of the paper focused on new tools for the assessment of sarcopenia with electromyography (EMG).*

yielded the best performance; showing that the signal of the inertial sensor contained abundant information on gait parameters. However, the deep learning did not extract effective features from inertial signals; further data and greater cohorts, respectively, with additional clinical evaluations should be collected and studied [14].
