Background: The ultrasound is one of the most used medical imaging investigations worldwide. It is non-invasive and effective in assessing liver tumors or other types of parenchymal changes. Methods: The aim of the study was to build a deep learning model for image segmentation in ultrasound video investigations. The dataset used in the study was provided by the University of Medicine and Pharmacy Craiova, Romania and contained 50 video examinations from 49 patients. The mean age of the patients in the cohort was 69.57. Regarding presence of a subjacent liver disease, 36.73% had liver cirrhosis and 16.32% had chronic viral hepatitis (5 patients: chronic hepatitis C and 3 patients: chronic hepatitis B). Frames were extracted and cropped from each examination and an expert gastroenterologist labelled the lesions in each frame. After labelling, the labels were exported as binary images. A deep learning segmentation model (U-Net) was trained with focal Tversky loss as a loss function. Two models were obtained with two different sets of parameters for the loss function. The performance metrics observed were intersection over union and recall and precision. Results: Analyzing the intersection over union metric, the first segmentation model obtained performed better compared to the second model: 0.8392 (model 1) vs. 0.7990 (model 2). The inference time for both models was between 32.15 milliseconds and 77.59 milliseconds. Conclusions: Two segmentation models were obtained in the study. The models performed similarly during training and validation. However, one model was trained to focus on hard-to-predict labels. The proposed segmentation models can represent a first step in automatically extracting time-intensity curves from CEUS examinations.

Background: Contrast-enhanced ultrasound (CEUS) is an important imaging modality in the diagnosis of liver tumors. By using contrast agent, a more detailed image is obtained. Time-intensity curves (TIC) can be extracted using a specialized software, and then the signal can be analyzed for further investigations. Methods: The purpose of the study was to build an automated method for extracting TICs and classifying liver lesions in CEUS liver investigations. The cohort contained 50 anonymized video investigations from 49 patients. Besides the CEUS investigations, clinical data from the patients were provided. A method comprising three modules was proposed. The first module, a lesion segmentation deep learning (DL) model, handled the prediction of masks frame-by-frame (region of interest). The second module performed dilation on the mask, and after applying colormap to the image, it extracted the TIC and the parameters from the TIC (area under the curve, time to peak, mean transit time, and maximum intensity). The third module, a feed-forward neural network, predicted the final diagnosis. It was trained on the TIC parameters extracted by the second model, together with other data: gender, age, hepatitis history, and cirrhosis history. Results: For the feed-forward classifier, five classes were chosen: hepatocarcinoma, metastasis, other malignant lesions, hemangioma, and other benign lesions. Being a multiclass classifier, appropriate performance metrics were observed: categorical accuracy, F1 micro, F1 macro, and Matthews correlation coefficient. The results showed that due to class imbalance, in some cases, the classifier was not able to predict with high accuracy a specific lesion from the minority classes. However, on the majority classes, the classifier can predict the lesion type with high accuracy. Conclusions: The main goal of the study was to develop an automated method of classifying liver lesions in CEUS video investigations. Being modular, the system can be a useful tool for gastroenterologists or medical students: either as a second opinion system or a tool to automatically extract TICs.

Background and Objectives. Bilateral knee osteoarthritis (KOA) in older patients compromises mobility and quality of life. Gait analysis provides objective, reproducible measures of physical performance. This study evaluated the integration of gait analysis for longitudinal monitoring and personalized rehabilitation, while ultrasound was performed only at baseline for characterization and did not inform adjustments to the interventions. Materials and Methods. We conducted a 6-week, randomized, parallel-group clinical trial including 130 participants aged ≥65 years. Patients were randomized to a Study Group (SG, n = 64) receiving 10 sessions of a comprehensive rehabilitation program (conventional measures plus gait training) or a Control Group (CG, n = 66) receiving 10 sessions of conventional rehabilitation (electrotherapy with magnetodiaflux, TENS, therapeutic ultrasound, low-intensity laser, plus standard kinesitherapy). Interventions were delivered daily, 5 days/week, over 2 consecutive weeks. Gait outcomes (BTS G-WALK/G-SENSOR 2) included TUG, Symmetry Index, 6MWD, and cadence; functional outcomes included VAS pain, WOMAC, and Lequesne Index. Quadriceps morphology was assessed sonographically, and a height-normalized quadriceps thickness index (QHNI) was calculated. Results. Of 130 randomized, 112 completed (93.3%). Compared with control, the intervention produced significant, clinically meaningful improvements: 6MWD increased by approximately 59 m, cadence by ~9 steps/min; TUG improved by ~2.6 s; gait symmetry by ~4–5 points; VAS pain decreased by ~1.7 points; WOMAC total by ~8.5 points; and Lequesne Index by ~2 points (all p < 0.001). QHNI showed no significant association with anthropometrics and performance measures, indicating limited value as a linear predictor at baseline; nonlinear models or subgroup analyses may be warranted. Conclusions. Both neuromuscular-focused rehabilitation and the conventional program improved gait and clinical outcomes. Integrating gait analysis with ultrasound evaluation enables comprehensive monitoring and supports personalized interventions to reduce joint loading and optimize gait mechanics in older patients with bilateral KOA.

Background: Sarcopenic obesity, characterized by reduced skeletal muscle mass and excess adiposity, is strongly associated with chronic pain and functional decline in older adults. Objective: This prospective controlled trial without randomization investigated the effects of a structured, three-phase rehabilitation program on physical performance, pain, and sarcopenia-specific quality of life in elderly patients with sarcopenic obesity and chronic pain. Methods: In this study, 82 participants were enrolled and allocated to a study group (SG, n = 40), receiving supervised rehabilitation, nutritional counseling, and supplementation, or to a control group (CG, n = 42), which did not receive rehabilitation. The final analysis included 35 patients in SG and 36 in CG. Outcomes were assessed at baseline and six months using the Sarcopenia Quality of Life Questionnaire (SarQoL), Short Physical Performance Battery (SPPB), Numeric Rating Scale (NRS), and Pressure Pain Threshold (PPT). Results: The SG showed significant improvements in all outcomes: SarQoL increased from 57.02 to 63.98, SPPB increased from 7.14 to 8.4, PPT increased from 69.31 to 78.05, and NRS decreased from 6.94 to 4.65 (all p < 0.001). The CG showed no significant changes. Conclusions: The implementation of a structured, three-phase rehabilitation program resulted in clinically and statistically significant improvements in physical performance, pain perception, and sarcopenia-related quality of life in older adults with sarcopenic obesity and chronic pain.

Sarcopenia, defined as a progressive loss of skeletal muscle mass, strength, and function, is a leading contributor to disability, dependence, and reduced quality of life (HRQoL) in older adults. This study aimed to evaluate the impact of a personalized six-month rehabilitation program, centered on tailored kinetic therapy, on physical performance and HRQoL in older women with primary sarcopenia. Methods: This prospective controlled study included 80 women aged ≥65 years, allocated into a Study Group (SG, n = 40), who followed a supervised personalized kinetic program, and a control group (CG, n = 40), who received general advice regarding physical activity and nutrition. Physical performance was measured using the short physical performance battery (SPPB), while HRQoL was assessed with the disease-specific SarQoL questionnaire. Evaluations were conducted at baseline and after six months. Results: At baseline, both groups had comparable scores (SPPB: SG = 5.75 ± 0.86 vs. CG = 5.8 ± 0.88, p = 0.798; SarQoL: SG = 54.42 ± 8.76 vs. CG = 55.59 ± 4.61, p = 0.457). After six months, the SG showed significant improvements (SPPB = 8.05 ± 0.90, p < 0.001; SarQoL = 62.55 ± 7.00, p < 0.001). Significant gains were observed in domains related to physical and mental health, locomotion, functionality, and leisure activities (p < 0.05). In contrast, the CG showed only minor, non-significant changes (SPPB = 6.17 ± 0.78; SarQoL = 56.51 ± 5.51). Conclusions: A structured, personalized kinetic program significantly improves physical performance and HRQoL in older women with primary sarcopenia. These results support the need for individualized, supervised rehabilitation programs in optimizing functional recovery and enhancing patient-centered outcomes in sarcopenia management.

This study aims to develop a scoring method that can be used by primary care physicians from remote areas or resource-limited settings to estimate the need for fecal occult blood testing (FOBT) as a first step in colorectal cancer screening. This method relies on several modifiable risk factors that can influence a positive FOBT, an indication of the presence of colorectal polyps, or even colorectal cancer. The scoring method considers, besides the age and gender of the patient, the body mass index (BMI), smoking status, and the diagnoses of diabetes mellitus (type 2 diabetes), dyslipidemia, and hypertension. It does not need any paraclinical exams, which is an advantage when access or material resources are limited. The retrospective study was spread over forty-three months, respectively, from October 2019 to April 2023, and included 112 patients. The score that we designed is a numerical value between 0 and 7. The values between 0 and 3 represent a smaller risk of a positive FOBT (9.68%), values 4 and 5 represent a medium risk (14.75%), while values 6 and 7 represent a greater risk (40%). Using this score, a physician can determine if a patient has a greater risk and recommend it to prioritize taking a FOB test.