In-memory factorization of holographic perceptual representations.

Disentangling the attributes of a sensory signal is central to sensory perception and cognition and hence is a critical task for future artificial intelligence systems. Here we present a compute engine capable of efficiently factorizing high-dimensional holographic representations of combinations of such attributes, by exploiting the computation-in-superposition capability of brain-inspired hyperdimensional computing, and the intrinsic stochasticity associated with analogue in-memory computing based on nanoscale memristive devices. Such an iterative in-memory factorizer is shown to solve at least five orders of magnitude larger problems that cannot be solved otherwise, as well as substantially lowering the computational time and space complexity. We present a large-scale experimental demonstration of the factorizer by employing two in-memory compute chips based on phase-change memristive devices. The dominant matrix-vector multiplication operations take a constant time, irrespective of the size of the matrix, thus reducing the computational time complexity to merely the number of iterations. Moreover, we experimentally demonstrate the ability to reliably and efficiently factorize visual perceptual representations.

From basic sciences and engineering to epileptology: A translational approach.

Collaborative efforts between basic scientists, engineers, and clinicians are enabling translational epileptology. In this article, we summarize the recent advances presented at the International Conference for Technology and Analysis of Seizures (ICTALS 2022): (1) novel developments of structural magnetic resonance imaging; (2) latest electroencephalography signal-processing applications; (3) big data for the development of clinical tools; (4) the emerging field of hyperdimensional computing; (5) the new generation of artificial intelligence (AI)-enabled neuroprostheses; and (6) the use of collaborative platforms to facilitate epilepsy research translation. We highlight the promise of AI reported in recent investigations and the need for multicenter data-sharing initiatives.

Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks.

Memory-augmented neural networks enhance a neural network with an external key-value (KV) memory whose complexity is typically dominated by the number of support vectors in the key memory. We propose a generalized KV memory that decouples its dimension from the number of support vectors by introducing a free parameter that can arbitrarily add or remove redundancy to the key memory representation. In effect, it provides an additional degree of freedom to flexibly control the tradeoff between robustness and the resources required to store and compute the generalized KV memory. This is particularly useful for realizing the key memory on in-memory computing hardware where it exploits nonideal, but extremely efficient nonvolatile memory devices for dense storage and computation. Experimental results show that adapting this parameter on demand effectively mitigates up to 44% nonidealities, at equal accuracy and number of devices, without any need for neural network retraining.

Evaluating the effective pressure applied by a valgus knee orthosis in individuals with medial knee osteoarthritis based on the dose-response relationship.

BACKGROUND: There is evidence that valgus knee orthosis improves clinical and biomechanical outcomes in individuals with medial knee osteoarthritis (MKOA). It is unclear whether variations in pressure application by orthosis straps can affect the biomechanical outcomes. This study aimed to determine the dose-response relationship between different orthosis straps tensions and changes in knee adduction moment (KAM) parameters in individuals with MKOA. METHOD: Twenty-four individuals with symptomatic MKOA were enrolled in this quasi-experimental study. Five tension conditions in orthosis straps were tested in 20-mmHg increments, from 0 (no pressure) to 100 (maximal pressure) mmHg. Patients were asked to adjust the orthosis strap tension based on their perceived comfort. After each condition, a 3D gait analysis was performed, and KAM parameters were calculated. The participants also reported their satisfaction with knee orthosis adjustment for each pressure condition. RESULTS: With successive increases in strap tension from 40 to 80 mmHg, the first peak, second peak, and angular impulse of KAM decreased nonlinearly (from 6 % to 25 %). Increasing the orthosis strap tension to 100 mmHg significantly decreased (P < 0.05) the participants' satisfaction level. The effective dosages (IC50) of pressure for the first peak, second peak, and angular impulse of KAM as responses were 58, 65, and 69 mmHg, respectively. CONCLUSION: The KAM decline was not linear as the strap pressure increased. Patients were dissatisfied with orthosis adjustment when strap tension was above 80 mmHg. The optimum dosage of pressure on the knee joint's lateral side for adjusting an orthosis' strap tension is approximately 69 mmHg.

Vector Symbolic Architectures as a Computing Framework for Emerging Hardware.

This article reviews recent progress in the development of the computing framework Vector Symbolic Architectures (also known as Hyperdimensional Computing). This framework is well suited for implementation in stochastic, emerging hardware and it naturally expresses the types of cognitive operations required for Artificial Intelligence (AI). We demonstrate in this article that the field-like algebraic structure of Vector Symbolic Architectures offers simple but powerful operations on high-dimensional vectors that can support all data structures and manipulations relevant to modern computing. In addition, we illustrate the distinguishing feature of Vector Symbolic Architectures, "computing in superposition," which sets it apart from conventional computing. It also opens the door to efficient solutions to the difficult combinatorial search problems inherent in AI applications. We sketch ways of demonstrating that Vector Symbolic Architectures are computationally universal. We see them acting as a framework for computing with distributed representations that can play a role of an abstraction layer for emerging computing hardware. This article serves as a reference for computer architects by illustrating the philosophy behind Vector Symbolic Architectures, techniques of distributed computing with them, and their relevance to emerging computing hardware, such as neuromorphic computing.

The efficacy of intramuscular electrical stimulation in the management of patients with myofascial pain syndrome: a systematic review.

INTRODUCTION: Myofascial pain syndrome (MPS) is one of the most common disorders causing chronic muscle pain. Almost one-third of patients with musculoskeletal complaints meet the MPS criteria. The aim of this study is to evaluate the effectiveness of intramuscular electrical stimulation (IMES) in patients with MPS through a systematic review method. METHODS: PubMed, Scopus, Embase, ProQuest, PEDro, Web of Science, and CINAHL were systematically searched to find out the eligible articles without language limitations from 1990 to December 30, 2020. All relevant randomized controlled trials that compared the effectiveness of IMES with sham-IMES, dry needling, or exercise therapy in patients with MPS were included. Full texts of the selected studies were critically appraised using Revised Cochrane risk-of-bias tool for randomized trials (RoB2). RESULTS: Six studies (out of 397) had met our inclusion criteria (involving 158 patients) and were entered to the systematic review. Outcome measures examined in these studies included pain, range of motion, pressure pain threshold, biochemical factors, disability, and amount of analgesic use. In the most studies, it has been shown that IMES is more effective than the control group in improving some outcome measurements such as pain. CONCLUSION: There is preliminary evidence from a few small trials suggesting the efficacy of IMES for the care of myofascial pain syndrome. The data support the conduct of larger trials investigating the efficacy of IMES.

Near-channel classifier: symbiotic communication and classification in high-dimensional space.

Brain-inspired high-dimensional (HD) computing represents and manipulates data using very long, random vectors with dimensionality in the thousands. This representation provides great robustness for various classification tasks where classifiers operate at low signal-to-noise ratio (SNR) conditions. Similarly, hyperdimensional modulation (HDM) leverages the robustness of complex-valued HD representations to reliably transmit information over a wireless channel, achieving a similar SNR gain compared to state-of-the-art codes. Here, we first propose methods to improve HDM in two ways: (1) reducing the complexity of encoding and decoding operations by generating, manipulating, and transmitting bipolar or integer vectors instead of complex vectors; (2) increasing the SNR gain by 0.2 dB using a new soft-feedback decoder; it can also increase the additive superposition capacity of HD vectors up to 1.7[Formula: see text] in noise-free cases. Secondly, we propose to combine encoding/decoding aspects of communication with classification into a single framework by relying on multifaceted HD representations. This leads to a near-channel classification (NCC) approach that avoids transformations between different representations and the overhead of multiple layers of encoding/decoding, hence reducing latency and complexity of a wireless smart distributed system while providing robustness against noise and interference from other nodes. We provide a use-case for wearable hand gesture recognition with 5 classes from 64 EMG sensors, where the encoded vectors are transmitted to a remote node for either performing NCC, or reconstruction of the encoded data. In NCC mode, the original classification accuracy of 94% is maintained, even in the channel at SNR of 0 dB, by transmitting 10,000-bit vectors. We remove the redundancy by reducing the vector dimensionality to 2048-bit that still exhibits a graceful degradation: less than 6% accuracy loss is occurred in the channel at - 5 dB, and with the interference from 6 nodes that simultaneously transmit their encoded vectors. In the reconstruction mode, it improves the mean-squared error by up to 20 dB, compared to standard decoding, when transmitting 2048-dimensional vectors.

A Primer on Hyperdimensional Computing for iEEG Seizure Detection.

A central challenge in today's care of epilepsy patients is that the disease dynamics are severely under-sampled in the currently typical setting with appointment-based clinical and electroencephalographic examinations. Implantable devices to monitor electrical brain signals and to detect epileptic seizures may significantly improve this situation and may inform personalized treatment on an unprecedented scale. These implantable devices should be optimized for energy efficiency and compact design. Energy efficiency will ease their maintenance by reducing the time of recharging, or by increasing the lifetime of their batteries. Biological nervous systems use an extremely small amount of energy for information processing. In recent years, a number of methods, often collectively referred to as brain-inspired computing, have also been developed to improve computation in non-biological hardware. Here, we give an overview of one of these methods, which has in particular been inspired by the very size of brains' circuits and termed hyperdimensional computing. Using a tutorial style, we set out to explain the key concepts of hyperdimensional computing including very high-dimensional binary vectors, the operations used to combine and manipulate these vectors, and the crucial characteristics of the mathematical space they inhabit. We then demonstrate step-by-step how hyperdimensional computing can be used to detect epileptic seizures from intracranial electroencephalogram (EEG) recordings with high energy efficiency, high specificity, and high sensitivity. We conclude by describing potential future clinical applications of hyperdimensional computing for the analysis of EEG and non-EEG digital biomarkers.

Thoracic and Lumbar Sagittal Spinal Curvature Adaptations between Elite Iranian Road and Speed Cyclists.

BACKGROUND: Despite existing some research on the effects of professional cycling on cyclist's spinal curvature angles, no research is available concentrating on possible differences in various types of professional cycling. Road (outdoor) and speed (indoor) cyclists have different anthropometric and predominant postures during cycling. The current study aims to investigate if cyclists with different types of cycling may have dissimilar spinal curvature adaptations. OBJECTIVE: Forty-eight male subjects, including 16 elite road cyclists, 16 elite sprint cyclists and 16 non-athletes as the control group were recruited in this non-experimental study. MATERIAL AND METHODS: In this cross-sectional study, a spinal mouse was used to measure the thoracic and lumbar curvature angles in standing position. RESULTS: The mean values of thoracic kyphosis and lumbar lordosis angles were found as follows 48.3±7.2º & -20.3±7.2º for elite road cyclists; 46.6±8.1º & -22.5±7.7º for elite sprint cyclists; and 37.5±4.1º & -19±6.3º for the control group. The results confirmed that both the road and sprint cyclists showed significantly more hyper-kyphosis posture in their thoracic region when compared to the control group (p<0.05). CONCLUSION: All road cyclists showed no significantly higher degrees of thoracic angle relative to the sprint cyclists (p>0.05). However, all the cyclists and the control groups showed a normal range of lumbar lordosis angle. As a conclusion, the results of this study confirmed that an elite cyclist may predispose these subjects to the risk of hyper-kyphotic posture. The road cycling may result in more hyper-kyphosis due to the longer time, spent in flexed position.

A Pediatric Parathyroid Carcinoma: An Unusual Clinical Presentation and Mini-review.

INTRODUCTION: Primary hyperparathyroidism (PHPT) is a rare condition in the pediatric population. Parathyroid carcinoma (PC) is a very uncommon cause of PHPT, accounting for < 1% of pediatric PHPT cases. It is challenging to distinguish between parathyroid adenoma (PA), the most common cause of PHPT, and PC. In this report, we described a young female who presented with a history of progressive limping and was finally diagnosed with PC. CASE PRESENTATION: A 15-year-old girl presented with progressive limping and bone pain for 8 years. She was referred by an orthopedic surgeon because of elevated intact parathyroid hormone (iPTH) for further evaluation. Physical examination revealed a large, firm, and non-tender neck mass, left hip tenderness, and limited range of motion. The initial biochemistry tests showed a borderline high calcium level of 10.8 mg/dl, an elevated iPTH level of 2876 pg/mL, and a decreased phosphorus level of 2.4 mg/dL. The 99mTechnetium (Tc) sestamibi scan displayed early intense activity in the right thyroid lobe persisting in the three-hour repeat scan, compatible with a parathyroid lesion. The patient underwent right-sided neck exploration and parathyroidectomy. Intraoperative and pathology findings confirmed the diagnosis of PC. Immunohistochemistry (IHC) staining revealed creatine kinase (CK) and CD31 in endothelial cells of the tumor. Ki67 staining was also positive in 2% - 3% of tumor cells. The whole exome sequencing (WES) study was negative for cell division cycle 73 (CDC73) and multiple endocrine neoplasia 1 (MEN1) genes. CONCLUSIONS: PC should be considered as a differential diagnosis of PHPT in the pediatric population, even in the presence of mild hypercalcemia.

Robust high-dimensional memory-augmented neural networks.

Traditional neural networks require enormous amounts of data to build their complex mappings during a slow training procedure that hinders their abilities for relearning and adapting to new data. Memory-augmented neural networks enhance neural networks with an explicit memory to overcome these issues. Access to this explicit memory, however, occurs via soft read and write operations involving every individual memory entry, resulting in a bottleneck when implemented using the conventional von Neumann computer architecture. To overcome this bottleneck, we propose a robust architecture that employs a computational memory unit as the explicit memory performing analog in-memory computation on high-dimensional (HD) vectors, while closely matching 32-bit software-equivalent accuracy. This is achieved by a content-based attention mechanism that represents unrelated items in the computational memory with uncorrelated HD vectors, whose real-valued components can be readily approximated by binary, or bipolar components. Experimental results demonstrate the efficacy of our approach on few-shot image classification tasks on the Omniglot dataset using more than 256,000 phase-change memory devices. Our approach effectively merges the richness of deep neural network representations with HD computing that paves the way for robust vector-symbolic manipulations applicable in reasoning, fusion, and compression.

An Ensemble of Hyperdimensional Classifiers: Hardware-Friendly Short-Latency Seizure Detection With Automatic iEEG Electrode Selection.

We propose a new algorithm for detecting epileptic seizures. Our algorithm first extracts three features, namely mean amplitude, line length, and local binary patterns that are fed to an ensemble of classifiers using hyperdimensional (HD) computing. These features are embedded into prototype vectors representing ictal (during seizures) and interictal (between seizures) brain states are constructed. These vectors can be computed at different spatial scales ranging from a single electrode up to many electrodes. This flexibility allows our algorithm to identify the electrodes that discriminate best between ictal and interictal brain states. We assess our algorithm on the SWEC-ETHZ iEEG dataset that includes 99 short-time iEEG seizures recorded with 36 to 100 electrodes from 16 drug-resistant epilepsy patients. Using k-fold cross-validation and all electrodes, our algorithm surpasses state-of-the-art algorithms yielding significantly shorter latency (8.81 s vs. 11.57 s) in seizure onset detection, and higher specificity (97.31% vs. 94.84%) and accuracy (96.85% vs. 95.42%). We can further reduce the latency of our algorithm to 3.74 s by allowing a slightly higher percentage of false alarms (2% specificity loss). Using only the top 10% of the electrodes ranked by our algorithm, we still maintain superior latency, sensitivity, and specificity compared to the other algorithms with all the electrodes. We finally demonstrate the suitability of our algorithm to deployment on low-cost embedded hardware platforms, thanks to its robustness to noise/artifacts affecting the signal, its low computational complexity, and the small memory-footprint on a RISC-V microcontroller.

Hyperdimensional Computing With Local Binary Patterns: One-Shot Learning of Seizure Onset and Identification of Ictogenic Brain Regions Using Short-Time iEEG Recordings.

OBJECTIVE: We develop a fast learning algorithm combining symbolic dynamics and brain-inspired hyperdimensional computing for both seizure onset detection and identification of ictogenic (seizure generating) brain regions from intracranial electroencephalography (iEEG). METHODS: Our algorithm first transforms iEEG time series from each electrode into symbolic local binary pattern codes, from which a holographic distributed representation of the brain state of interest is constructed across all the electrodes and over time in a hyperdimensional space. The representation is used to quickly learn from few seizures, detect their onset, and identify the spatial brain regions that generated them. RESULTS: We assess our algorithm on our dataset that contains 99 short-time iEEG recordings from 16 drug-resistant epilepsy patients being implanted with 36-100 electrodes. For the majority of the patients (ten out of 16), our algorithm quickly learns from one or two seizures and perfectly (100%) generalizes on novel seizures using k-fold cross-validation. For the remaining six patients, the algorithm requires three to six seizures for learning. Our algorithm surpasses the state-of-the-art including deep learning algorithms by achieving higher specificity (94.84% versus 94.77%) and macroaveraging accuracy (95.42% versus 94.96%), and 74× lower memory footprint, but slightly higher average latency in detection (15.9 s versus 14.7 s). Moreover, the algorithm can reliably identify (with a p-value ) the relevant electrodes covering an ictogenic brain region at two levels of granularity: cerebral hemispheres and lobes. CONCLUSION AND SIGNIFICANCE: Our algorithm provides: 1) a unified method for both learning and classification tasks with end-to-end binary operations; 2) one-shot learning from seizure examples; 3) linear computational scalability for increasing number of electrodes; and 4) generation of transparent codes that enables post-translational support for clinical decision making. Our source code and anonymized iEEG dataset are freely available at http://ieeg-swez.ethz.ch.

The effects of chromium and vitamin D3 co-supplementation on insulin resistance and tumor necrosis factor-alpha in type 2 diabetes: a randomized placebo-controlled trial.

The current study was conducted to assess the effects of simultaneous usage with vitamin D3 and chromium picolinate (CrPic) supplementations on homeostasis model assessment of insulin resistance (HOMA-IR), fasting blood glucose (FBS), hemoglobin A1c (HbA1c), tumor necrosis factor-α (TNF-α), and lipid profile in type 2 diabetes mellitus (T2DM). Ninety-two patients with T2DM were randomly allocated to the following 4 groups for 4 months: (I) placebo of vitamin D3 (n = 23); (II) vitamin D3 supplement at a dose of 50 000 IU/week (n = 23); (III) CrPic supplement at a dose of 500 µg/day (n = 23); and (IV) both vitamin D3 at a dose of 50 000 IU/week and CrPic at a dose of 500 µg/day (n = 23). HOMA-IR levels increased significantly in groups I and II after the intervention. However, this increase in group I was significantly higher than that in group II after the treatment. HOMA-IR levels were controlled in groups III and IV during the intervention. TNF-α decreased significantly in groups II, III, and IV after the intervention. FBS, HbA1c, and lipid profile did not change significantly in total groups after the intervention. It seems that chromium and vitamin D3 co-supplementation are probably effective in controlling HOMA-IR by decreasing TNF-α in T2DM. Novelty Chromium alone and/or in simultaneous pretreatment with vitamin D3 is more effective than vitamin D3 in controlling HOMA-IR in T2DM. Chromium and vitamin D3 alone and/or in simultaneous pretreatment decrease TNF-α in T2DM.

Studying the correlation between balance assessment by Biodex Stability System and Berg Scale in stroke individuals.

INTRODUCTION: Balance disorders are considered to be a serious clinical manifestation after stroke. Therefore, to assess stroke patients' balance performance, use of a quantitative method appears essential. A fundamental step would be the approval of the efficiency of the measurement instruments. The current study aimed to investigate correlations between balance assessment as examined by Biodex Stability System (BSS) and the clinical Berg Balance Scale (BBS) in post-stroke hemiparesis. METHODS: Twenty-five stroke survivors and 25 healthy age-sex matched subjects were recruited. The subjects were assessed using BSS during 3 days, with a 24-h interval. The high interclass correlation coefficient (ICC) values showed that the system was reliable enough to continue the study. The clinical evaluation was performed by the standard BBS. RESULTS: There was a significant moderate negative correlation between the Biodex overall indices and BBS scores in the stroke groups (ravg = -0.68) and in the healthy cohort (ravg = -0.55). Also, a significant moderate negative correlation was found between the Biodex antero-posterior stability indices and BBS scores in the stroke groups (ravg = -0.67) and in healthy cohort (ravg = -0.55). The correlation between the Biodex mediolateral stability indices and BBS scores was moderate to low in the stroke and healthy groups (ravg = -0.67 and -0.39 respectively). DISCUSSION AND CONCLUSION: Moderate negative correlation between the stability indices of the Biodex Stability System and BBS scores indicates that dynamic balance status of the participants partially reflects their functional balance status.

Online Learning and Classification of EMG-Based Gestures on a Parallel Ultra-Low Power Platform Using Hyperdimensional Computing.

This paper presents a wearable electromyographic gesture recognition system based on the hyperdimensional computing paradigm, running on a programmable parallel ultra-low-power (PULP) platform. The processing chain includes efficient on-chip training, which leads to a fully embedded implementation with no need to perform any offline training on a personal computer. The proposed solution has been tested on 10 subjects in a typical gesture recognition scenario achieving 85% average accuracy on 11 gestures recognition, which is aligned with the state-of-the-art, with the unique capability of performing online learning. Furthermore, by virtue of the hardware friendly algorithm and of the efficient PULP system-on-chip (Mr. Wolf) used for prototyping and evaluation, the energy budget required to run the learning part with 11 gestures is 10.04 mJ, and 83.2  µJ per classification. The system works with a average power consumption of 10.4 mW in classification, ensuring around 29 h of autonomy with a 100 mAh battery. Finally, the scalability of the system is explored by increasing the number of channels (up to 256 electrodes), demonstrating the suitability of our approach as universal, energy-efficient biopotential wearable recognition framework.

Effect of using the Flexi-Bar tool on erector spinae muscle activation under different standing weight-bearing conditions.

OBJECTIVES: The Flexi-Bar and Bodyblade are oscillatory tools used in rehabilitation centers worldwide to enhance muscle activity. Because of a lack of reports on the Flexi-Bar, this study focused on erector spinae (ES) muscle activation under different conditions. METHODS: Twelve university students (age 21 ± 2.5 years old) were recruited in this study and were tested while using a loaded Flexi-Bar for 10 seconds. Comparison between muscle activation on the right and left sides in the cervical, thoracic, and lumbar regions was measured by electromyography during two-leg and one-leg (left and right) standing. RESULTS: The results showed that during oscillation using the right hand, the right cervical muscles showed significantly higher activation levels than the left cervical muscles, while the thoracic and lumbar regions showed significantly higher muscle activities on the left side. CONCLUSION: The current study showed that the Flexi-Bar might be used for muscle activation and reeducation, as well as for specific exercise therapy in spinal muscle imbalance.

Electromyographic activity of the hip and knee muscles during functional tasks in males with and without patellofemoral pain.

BACKGROUND: Patellofemoral pain (PFP) is a common overuse injury in physically active individuals. It is characterized by anterior knee, retropatellar, or prepatellar pain associated with activities that increase patellofemoral joint stress such as squatting, stair ascending and descending, running, jumping, prolonged sitting, and kneeling. The etiology of PFP is believed to be multifactorial. Recently, proximal factors have been shown to influence the biomechanics of patellofemoral joint. OBJECTIVE: The aim of the study was to assess hip and knee muscle activity during single leg stance and single leg squat in males with PFP and a control group without PFP. METHODS: Eighteen males with PFP (age 24.2 ±â€¯4.4 years) and 18 healthy subjects as controls (age 23.5 ±â€¯3.8 years) were included. We evaluated gluteus medius, gluteus maximus, vastus medialis oblique (VMO), and vastus lateralis (VL) electromyographic (EMG) activity. The muscle activity and reaction time of the proposed muscles were assessed during single leg stance and single leg squat tasks. Independent t-test was used to identify significant differences between PFP and control groups. RESULTS: No difference in activity of the gluteus maximus muscle was found in either task (p > 0.5). Significant differences were found in activity of gluteus medius and VMO in both tasks (p < 0.05). VL muscle activity had significant difference in single leg stance (p = 0.01), however, had no significant difference in single leg squat (p = 0.1). No significant differences were found in reaction time of the four studied muscles during both single leg stance and single leg squat (p > 0.5). CONCLUSION: Males with PFP demonstrated altered gluteus medius, VMO, and VL muscle activity during single leg stance and single leg squat compared to healthy subjects. Gluteus maximus activity did not show any changes between groups. Moreover, muscle recruitment patterns were different between PFP and healthy groups.

Are Twenty-Four Sessions of Aerobic Exercise Sufficient for Improving Cardiac Parameters in Diabetes Mellitus? A Randomized Controlled Trial.

Background: Diabetes is a chronic disease that reduces cardiorespiratory fitness and increases systolic and diastolic blood pressures as well as resting heart rate due to the activity level of the sympathetic nervous system. The aim of this study was to assess the effectiveness of 2 types of aerobic exercise, with and without external loading, on cardiac parameters in diabetic patients. Methods: This randomized controlled trial was carried out on 45 volunteers. These individuals were randomly divided into aerobic, weighted vest, and control groups. The aerobic protocol comprised 24 sessions of aerobic exercise. The exercise program for the weighted vest group was identical to that of the aerobic group, except that the subjects wore a weighted vest. The parameters were measured before and after the 24 sessions. Results: The mean age of the study population was 48.30±5.02 years in the aerobic group, 48.33±5.74 years in the weighted vest group, and 48.60±4.79 years in the control group. Males comprised 7 (53.8%) patients in the aerobic group, 7 (58.3%) in the weighted vest group, and 8 (53.3%) in the control group. After 8 weeks, maximum oxygen consumption in the aerobic group (mean±SD=37.54±8.02 mL/kg/min, 95% CI: 5.48 to 11.60; P<0.001) and the weighted vest group (mean±SD=35.92±3.96 mL/kg/min, 95% CI: 4.36 to 9.64; P<0.001) was increased, similar to metabolic equivalent of task in the aerobic group (mean±SD=11.60±1.62 kcal/kg×h, 95% CI: 1.48 to 2.72; P<0.001) and the weighted vest group (mean±SD=11.21±1.11 kcal/kg×h, 95% CI: 1.23 to 2.28; P<0.001). Furthermore, resting heart rate decreased significantly in the aerobic group (mean ± SD=90.23±8.90 bpm, 95% CI: -13.93 to -1.29; P=0.022) and the weighted vest group (mean±SD=90.58±9.19 bpm, 95% CI: -0.16 to - 12.33; P=0.045). Conclusion: These findings suggest that 24 aerobic exercise sessions might improve cardiac parameters in type 2 diabetes.