Journal of Biomedical Physics and Engineering最新文献

Background: High-Intensity Focused Ultrasound (HIFU) represents a non-invasive treatment approach that utilizes non-ionizing radiation. This technique has found clinical utility in managing both benign and malignant solid tumors.

Objective: This study aimed to investigate the variations in HIFU frequency and duty cycle influence thermal lesion formation in tissue to identify the optimal parameter combination for HIFU therapy in multi-layered tissues.

Material and methods: In this theoretical framework, a model of HIFU application to multi-layer biological tissues was created. Four unique HIFU parameter sets, defined by combining high or low frequency with high or low duty cycle, were comprehensively examined. The study analyzed how these settings influenced temperature distributions and lesion area in the layered tissue to ascertain the ideal combination of frequency and duty cycle parameters.

Results: Simulation results revealed that the former parameter set (high frequency, low duty cycle) was optimal for treating smaller, superficial tumours, whereas the latter combination (low frequency, high duty cycle) proved effective for deeper-seated lesions. Regarding thermal dose metrics, the high-energy setting (high frequency, high duty cycle) generated the most extensive lesion area and highest peak temperature, in contrast to the low-energy configuration (low frequency, low duty cycle), which produced the smallest coagulation zone and lowest focal temperature.

Conclusion: The study demonstrates that optimal HIFU therapeutic outcomes require frequency-duty cycle combinations tailored to tumour characteristics, with high-frequency/low-duty cycle for shallow small tumours and low-frequency/high-duty cycle for deep lesions, providing a framework for precision parameter selection in clinical applications.

Background: Wireless Capsule Endoscopy (WCE) is the gold standard for painless and sedation-free visualization of the Gastrointestinal (GI) tract. However, reviewing WCE video files, which often exceed 60,000 frames, can be labor-intensive and may result in overlooking critical frames. A proficient diagnostic system should offer gastroenterologists high sensitivity and Negative Predictive Value (NPV) to enhance diagnostic accuracy.

Objective: The current study aimed to establish a reliable expert diagnostic system using a hybrid classification approach, acknowledging the limitations of individual deep learning models in accurately classifying prevalent GI lesions. Introducing a hybrid classification framework, ensemble learning techniques were applied to Deep Convolutional Neural Networks (DCNNs) tailored for WCE frame analysis.

Material and methods: In this analytical study, DCNN models were trained on balanced and unbalanced datasets and then applied for classification. A model scoring hybrid classification approach was used to create meta-learners from the DCNN classifiers. Class scoring was utilized to refine decision boundaries for each class within the hybrid classifiers.

Results: The VG_BFCG model, constructed on a pre-trained VGG16, demonstrated robust classification performance, achieving a recall of 0.952 and an NPV of 0.977. Tuned hybrid classifiers employing class scoring outperformed model scoring counterparts, attaining a recall of 0.988 and an NPV of 1.00, compared to 0.979 and 0.989, respectively.

Conclusion: The unbalanced dataset, with a higher number of Angiectasia frames, enhanced the classification metrics for all models. The findings of this study underscore the crucial role of class scoring in improving the classification metrics for multi-class hybrid classification.

Background: Employing 2D rebinned sinograms in PET scanners has the potential to accelerate the overall reconstruction speed. Among the available rebinning techniques, Single-Slice Rebinning (SSRB) offers a computationally efficient approach.

Objective: This study aimed to evaluate the influence of varying span and Maximum Ring Difference (MRD) parameters in SSRB on the image quality of the Xtrim PET scanner.

Material and methods: This Monte Carlo simulation study used a GATE-simulated Xtrim-PET scanner. 3D list-mode data were histogrammed into 576 sinograms, and SSRB was applied to generate 2D sinograms. Subsequently, Maximum-Likelihood Expectation-Maximization (MLEM) reconstruction was performed on the sinograms with different MRD and span. Image quality was assessed using image quality, rod, and uniform phantoms. Furthermore, axial resolution was evaluated using point sources.

Results: Analysis of linear profiles in uniform phantom revealed a 2.6 mm inaccuracy in axial activity estimation when comparing spans of 21 and 7. Increased span and MRD lead to artifactual data in regions of high activity gradients, as observed in both uniform and rod phantoms. However, the Recovery Coefficient (RC) and Spilled-Over Ratio (SOR) remained unaffected. Concomitantly, increasing the span improved uniformity and reduced the coefficient of variation by 1.6% and 5.9%, respectively. Axial resolution remained largely unaffected by variations in span and MRD.

Conclusion: The RC and SOR remain robust to variations in span and MRD. However, higher levels of axial data compression were associated with the introduction of axial artifacts. Additionally, axial resolution was unaffected by increases in span and MRD, likely due to the limited field of view of the Xtrim-PET scanner.

The increasing prevalence of cardiovascular diseases underscores the need for efficient and user-friendly tools to monitor heart health. Traditional Holter monitors, while effective, are often bulky and inconvenient, limiting their use in real-world scenarios. This study introduces the Smart Portable Holter, a wireless device designed for real-time cardiac monitoring, enabling early detection of heart irregularities with enhanced accuracy and user convenience. The device captures continuous electrocardiogram signals and transmits them to a secure cloud platform for processing. Machine learning models, including Random Forest and Extreme Gradient Boosting (XGBoost), analyze the data to detect cardiac events. The system's performance was evaluated using real-world datasets, emphasizing accuracy and reliability in identifying cardiac arrhythmias. The Smart Portable Holter delivers an impressive 98% accuracy in detecting cardiac events. Its compact and wireless design enhances user comfort, allowing for seamless wear throughout the day. Coupled with advanced analytics, it offers detailed, time-stamped records that empower both users and healthcare professionals. These features facilitated early diagnosis and supported personalized treatment planning for patients with varying cardiac conditions. The Smart Portable Holter represents a significant advancement in cardiac care, combining portability, real-time analytics, and high diagnostic accuracy. By empowering patients and healthcare providers with actionable insights, it fosters proactive heart health management and contributes to improved clinical outcomes.

Chronic pain is a prevalent condition and the leading cause of work absenteeism worldwide. This condition involves persistent pain lasting more than three months, significantly impacting the quality of life and social interactions of patients. While the causes of chronic pain can often remain unknown, no definitive cure exists for the various known causes. Furthermore, the evaluation and prediction of pain can be challenging, particularly in unconscious patients receiving care in the intensive care unit. Subjective measures and traditional methods are typically employed for diagnosis, assessment, and treatment to identify the most effective approach. However, recent advancements in Artificial Intelligence (AI) and other computer science fields have revolutionized the medical domain, offering a novel and promising avenue for enhancing pain management. This review provides an overview of the potential benefits, limitations, and prospects associated with the role of AI in the diagnosis, assessment, and management of chronic pain.

Background: Cancer has emerged as a critical global health concern due to its widespread prevalence and impact on individuals, families, communities, and healthcare systems worldwide.

Objective: We investigated the anticancer effectiveness of capecitabine (CAP) and vorinostat (VOR) when incorporated into self-nanoemulsifying drug delivery systems (SNEDDSs).

Material and methods: In this experimental study, the SNEDDSs were formulated using polyethylene glycol 600 (PEG 600), castor oil and Tween 80. A ternary phase diagram was plotted for the SNEDDSs components and the single-phase formation region was attained. SNEDDSs were then prepared by dilution of the selected ratios of these components in water. Blank SNEDDSs containing ratios (in weight) of castor oil:Tween 80:PEG 600 of 50:30:20 (S1-SNEDDS) and 25:15:60 (S2-SNEDDS) were selected. S1-SNEDDS was loaded with CAP (S1-SNEDDS-CAP), and S2-SNEDDS was loaded with VOR (S2-SNEDDS-VOR).

Results: The developed SNEDDSs formed oil nanodroplets without phase separation. Using dynamic laser light scattering, S1-SNEDDS, S2-SNEDDS, S1-SNEDDS-CAP and S2-SNEDDS-VOR had droplets with average sizes of 171±37, 82±18, 117±26 and 37±8 nm, respectively, accompanied by span values of 0.96, 0.95, 0.97 and 0.96, respectively. CAP and VOR were effectively loaded into the SNEDDSs with high entrapment efficiencies and loading capacities. Considerable improvements in cells viability for CAP and VOR were attained upon loading into SNEDDSs. TUNEL assays of the cells upon treatment by S1-SNEDDS-CAP and S2-SNEDDS-VOR revealed a significant apoptosis in all the cells.

Conclusion: The study provides valuable insights into the potential of utilizing SNEDDSs as a novel delivery system for improving the anticancer properties of CAP and VOR.

Background: Modern radiotherapy techniques can destroy tumors with less harm to surrounding normal tissues. Normal Tissue Complication Probability (NTCP) models are useful to evaluate treatment plans.

Objective: This study aimed to use the Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) program to evaluate dose-volume indicators and radiobiological parameters for complications of the rectum and bladder in prostate cancer patients undergoing pelvic radiotherapy.

Material and methods: In this retrospective cross-sectional study, treatment planning information was gathered from 35 patients with pelvic lymph node involvement. Of these, 17 and 18 were treated using the three-dimensional Conformal Radiotherapy Technique (3D-CRT) and the Helical Tomotherapy (HT) technique, respectively. The Lyman-Kutcher-Burman and Relative Seriality models were used in conjunction with dose-volume histograms to calculate the NTCP values for the rectum and bladder.

Results: In the HT group compared to the 3D-CRT group, the values of D-Mean, V-40, V-50, V-60, and V-65 were lower for both the rectum and bladder. The NTCP values for grade 2 rectal bleeding, proctitis, and bladder toxicity were lower in the HT group. The dose-volume data of 67% of the HT patients satisfied all QUANTEC criteria, while only 30% of the 3D-CRT those met criteria.

Conclusion: The QUANTEC criteria were satisfied for the rectum and bladder in the HT and 3D-CRT groups, except for V-50, V-60, and V-65 of the rectum in 3D-CRT patients. The NTCP values for both organs were lower in the HT group than in the 3D-CRT group.

Background: Accurate bone age assessment is essential for determining the actual degree of development and indicating a disorder in growth. While clinical bone age assessment techniques are time-consuming and prone to inter/intra-observer variability, deep learning-based methods are used for automated bone age estimation.

Objective: The current study aimed to develop an unsupervised pre-training approach for automatic bone age estimation, addressing the challenge of limited labeled data and unique features of radiographic images of hand bones. Bone age estimation is complex and usually requires more labeling data. On the other hand, there is no model trained with hand radiographic images, reused for bone age estimation.

Material and methods: In this fundamental-applied research, the collection of Radiological Society of North America (RSNA) X-ray image collection is used to evaluate the efficiency of the proposed bone age estimation method. An autoencoder is trained to reconstruct the original hand radiography images. Then, a model based on the trained encoder produces the final estimation of bone age.

Results: Experimental results on the Radiological Society of North America (RSNA) X-ray image collection achieve a Mean Absolute Error (MAE) of 9.3 months, which is comparable to state-of-the-art methods.

Conclusion: This study presents an approach to estimating bone age on hand radiographs utilizing unsupervised pre-training with an autoencoder and also highlights the significance of autoencoders and unsupervised learning as efficient substitutes for conventional techniques.

Background: The characteristics of medical data in Precision Medicine (PM), the challenges related to their storage and retrieval, and the effective facilities to address these challenges are importantly considered in implementing PM. For this purpose, a secured and scalable infrastructure for various data integration and storage is needed.

Objective: This study aimed to determine the characteristics of PM data and recognize the challenges and solutions related to appropriate infrastructure for data storage and its related issues.

Material and methods: In this systematic study, coherent research was conducted on Web of Science, Scopus, PubMed, Embase, and Google Scholar from 2015 to 2023. A total of 16 articles were selected and evaluated based on the inclusion and exclusion criteria and the central search theme of the study.

Results: A total of 1,961 studies were identified from designated databases, 16 articles met the eligibility criteria and were classified into five main sections PM data and its major characteristics based on the volume, variety and velocity (3Vs) of medical big data, data quality issues, appropriate infrastructure for PM data storage, cloud computing and PM infrastructure, and security and privacy. The variety of PM data is categorized into four major categories.

Conclusion: A suitable infrastructure for precision medicine should be capable of integrating and storing heterogeneous data from diverse departments and sources. By leveraging big data management experiences from other industries and aligning their characteristics with those in precision medicine, it is possible to facilitate the implementation of precision medicine while avoiding duplication.