Diagnostic test accuracy of machine learning algorithms for the detection intracranial hemorrhage: a systematic review and meta-analysis study

BACKGROUND

This systematic review and meta-analysis were conducted to objectively evaluate the evidence of machine learning (ML) in the patient diagnosis of Intracranial Hemorrhage (ICH) on computed tomography (CT) scans.

METHODS

Until May 2023, systematic searches were conducted in ISI Web of Science, PubMed, Scopus, Cochrane Library, IEEE Xplore Digital Library, CINAHL, Science Direct, PROSPERO, and EMBASE for studies that evaluated the diagnostic precision of ML model-assisted ICH detection. Patients with and without ICH as the target condition who were receiving CT-Scan were eligible for the research, which used ML algorithms based on radiologists' reports as the gold reference standard. For meta-analysis, pooled sensitivities, specificities, and a summary receiver operating characteristics curve (SROC) were used.

RESULTS

At last, after screening the title, abstract, and full paper, twenty-six retrospective and three prospective, and two retrospective/prospective studies were included. The overall (Diagnostic Test Accuracy) DTA of retrospective studies with a pooled sensitivity was 0.917 (95% CI 0.88-0.943, I2 = 99%). The pooled specificity was 0.945 (95% CI 0.918-0.964, I2 = 100%). The pooled diagnostic odds ratio (DOR) was 219.47 (95% CI 104.78-459.66, I2 = 100%). These results were significant for the specificity of the different network architecture models (p-value = 0.0289). However, the results for sensitivity (p-value = 0.6417) and DOR (p-value = 0.2187) were not significant. The ResNet algorithm has higher pooled specificity than other algorithms with 0.935 (95% CI 0.854-0.973, I2 = 93%).

CONCLUSION

This meta-analysis on DTA of ML algorithms for detecting ICH by assessing non-contrast CT-Scans shows the ML has an acceptable performance in diagnosing ICH. Using ResNet in ICH detection remains promising prediction was improved via training in an Architecture Learning Network (ALN).

Particular issues to be considered in small field dosimetry for TrueBeam STx commissioning

This study aims to report the relevant issues concerning small fields in the commissioning of a TrueBeam STx for photon energies of 6MV, 10MV, 6FFF, and 10FFF. Percent depth doses, profiles, and field output factors were measured according to the beam model configuration of the treatment planning system. Multiple detectors were used based on the IAEA TRS-483 protocol as well as EBT3 radiochromic film. Analytical Anisotropic and Acuros XB algorithms, were configured and validated through basic dosimetry comparisons and end-to-end clinical tests.

Evaluating the Gray Level Co-Occurrence Matrix-Based Texture Features of Magnetic Resonance Images for Glioblastoma Multiform Patients' Treatment Response Assessment

Background

Medical images of cancer patients are usually evaluated qualitatively by clinical specialists which makes the accuracy of the diagnosis subjective and related to the skills of clinicians. Quantitative methods based on the textural feature analysis may be useful to facilitate such evaluations. This study aimed to analyze the gray level co-occurrence matrix (GLCM)-based texture features extracted from T1-axial magnetic resonance (MR) images of glioblastoma multiform (GBM) patients to determine the distinctive features specific to treatment response or disease progression.

Methods

20 GLCM-based texture features, in addition to mean, standard deviation, entropy, RMS, kurtosis, and skewness were extracted from step I MR images (obtained 72 h after surgery) and step II MR images (obtained three months later). Responded and not responded patients to treatment were classified manually based on the radiological evaluation of step II images. Extracted texture features from Step I and Step II images were analyzed to determine the distinctive features for each group of responsive or progressive diseases. MATLAB 2020 was applied to feature extraction. SPSS version 26 was used for the statistical analysis. P value < 0.05 was considered statistically significant.

Results

Despite no statistically significant differences between Step I texture features for two considered groups, almost all step II extracted GLCM-based texture features in addition to entropy M and skewness were significantly different between responsive and progressive disease groups.

Conclusions

GLCM-based texture features extracted from MR images of GBM patients can be used with automatic algorithms for the expeditious prediction or interpretation of response to the treatment quantitatively besides qualitative evaluations.

Pelvic radiation dose measurement for trauma patients in multifield radiographic examinations: A phantom-based TLD dosimetry study

Background and Aims

Trauma patients often suffer from multiple injuries and require undergoing various radiography which is referred to as multifield radiographic examinations. Protective measures may be ignored for these examinations due to stressful emergency situations or patients' conditions. This study was conducted to evaluate the scattered doses received by the pelvis during different common multifield radiographic examinations with an emphasis on field size adjustment.

Methods

A whole-body phantom, PBU-50, resembling the body mass, was used to carry out the common examinations for trauma patients (extremities, skull, chest, abdomen, pelvis, femur, and lumbar radiography), using a Pars Pad X-ray machine. To measure the primary entrance skin doses, three calibrated GR 200 thermoluminescence dosimeter (TLD) chips were placed in the central X-ray beam of scanned organs. Three TLDs were also placed on the pelvis symphysis pubis to measure the scattered dose received by the pelvis due to each carried-out radiography for standard and clinically used field sizes. A Harshaw 3500 TLD Reader was used to read the chips. TLD readouts (nano-Coulomb) were converted to dose (milli Gray [mGy]) using the predefined calibration curve.

Results

The scattered doses to the pelvis due to scanning a single organ differed from 0.80 to 1.70, and 0.82 to 4.09 mGy for standard and clinically used field sizes, respectively. The scattered doses to the pelvis in multifield examinations varied from 0.80 to 8.43 and 0.82 to 13.6 mGy for standard and clinically used field sizes, respectively, depending on the number of scanned organs and their distances from the pelvis.

Conclusions

Multiple and repeated radiographs combined with insufficient protective measures can increase the patient's dose. The findings indicate that the scattered doses received by the pelvis can exceed the reference values in multifield radiography, especially if the radiation field is not restricted properly to the scanned organ.

Evaluating the applications and effectiveness of magnetic nanoparticle-based hyperthermia for cancer treatment: A systematic review

Magnetic nanoparticle-based hyperthermia as a new cancer treatment technology has been applied for some kinds of tumors. To review the different applications and effectiveness of this new cancer treatment technique, PubMed, Science Direct, Web of Science, and Google Scholar databases were explored up to November 2022, using the following keywords combined in different ways: "Magnetic Nanoparticles Based Hyperthermia", "Magnetic Nanoparticles" AND "Hyperthermia" AND "Cancer". The obtained results were screened for the title and abstract and the relevant papers were reviewed for further details. Finally, 24 papers were included in the study. These papers have evaluated the application of magnetic nanoparticle-based hyperthermia for treating different cancers including breast, liver, prostate, pancreas, colon, brain, lung, and stem cell. Various nanoparticles including Iron Oxide (Fe₂O₃, Fe₃O₄), Dextran Spermine, Iron Chloride, Magnetic nanoparticles conjugated with Liposomes (MCLs), and Variable Molecular Weight Nanoparticles (VMWNPs) were used in different reviewed studies. The results of reviewed studies revealed that the nanoparticle-based hyperthermia technique as a new progressive modality can significantly improve treatment outcomes for some special cancers. Increasing life expectancy by up to 30% using Iron Oxide magnetic nanoparticle-based hyperthermia for pancreatic cancer and increasing tumor ablation by about 33% for other cancers were reported in reviewed articles. However, further studies are required to extend this new treatment technique to other cancers and for providing more accurate information on nanoparticle-based hyperthermia's effectiveness as a complementary technique in cancer treatment.

Evaluating the effects of variation in CT scanning parameters on the image quality and Hounsfield units for optimization of dose in radiotherapy treatment planning: A semi-anthropomorphic thorax phantom study

Aim

The diagnosis accuracy of computed tomography (CT) systems and the reliability of calculated Hounsfield Units (HUs) are critical in tumor detection and cancer patients' treatment planning. This study evaluated the effects of scan parameters (Kilovoltage peak or kVp, milli-Ampere-second or mAS reconstruction kernels and algorithms, reconstruction field of view, and slice thickness) on image quality, HUs, and the calculated dose in the treatment planning system (TPS).

Materials and Methods

A quality dose verification phantom was scanned several times by a 16-slice Siemens CT scanner. The DOSIsoft ISO gray TPS was applied for dose calculations. The SPSS.24 software was used to analyze the results and the P-value <0.05 was considered significant.

Results

Reconstruction kernels and algorithms significantly affected noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). The noise increased and CNR decreased by raising the sharpness of reconstruction kernels. SNR and CNR had considerable increments at iterative reconstruction compared with the filtered back-projection algorithm. The noise decreased by raising mAS in soft tissues. Also, KVp had a significant effect on HUs. TPS--calculated dose variations were less than 2% for mediastinum and backbone and less than 8% for rib.

Conclusions

Although HU variation depends on image acquisition parameters across a clinically feasible range, its dosimetric impact on the calculated dose in TPS can be neglected. Hence, it can be concluded that the optimized values of scan parameters can be applied to obtain the maximum diagnostic accuracy and calculate HUs more precisely without affecting the calculated dose in the treatment planning of cancer patients.

Follow‐up the severity of abnormalities diagnosed in chest CT imaging of COVID‐19 patients: A cross‐sectional study

Background and Aims

Methods

Results

Conclusion

Evaluation of the Scan Duration and Mechanical and Thermal Indices Applied for the Diagnostic Ultrasound Examinations

OBJECTIVE

Although epidemiological studies have failed to show the harmful effects of ultrasound in humans, as a form of energy, ultrasound has the potential to cause bio-effects in tissues. Therefore, clinical guidelines have been established for ultrasound technology related to human safety, which include Thermal (TI) and mechanical (MI) indices. The appropriate TI and MI ranges for embryonic examinations are between 0-1.0 and 0-0.4, respectively. The accepted TI and MI ranges are 0-2.0 and 0-1.9, respectively, for general ultrasound examinations. In addition, the scan duration should be kept as low as possible. Therefore, the present study aimed at evaluating the scan duration, TI, and MI as measures of acoustic output during ultrasound studies.

METHODS

A cross-sectional descriptive study was conducted for patients undergoing pregnancy checkups, routine checkups, and initial diagnosis ultrasound examinations. Samples were selected from imaging departments of 4 educational hospitals based on convenience sampling and 321 checklists completed by direct observation of ultrasound examinations.

RESULTS

For pregnancy scans, the mean TI and MI were obtained as 0.32 ± 0.27 and 1.15 ± 0.13, respectively. For non-pregnancy examinations, the mean value of TI and MI were 0.30 ± 0.29 and 1.07 ± 0.35, respectively. Therefore, mean TI for pregnancy and non-pregnancy examinations and mean MI for non-pregnancy studies obtained lower than the permitted values, while the mean MI was higher than the permitted level for first trimester of pregnancy. Also, relatively suitable scan durations were seen in reviewed studies.

CONCLUSION

From this study, it may be concluded that the reported ultrasound scans were safe.

Evaluating the principles of radiation protection in diagnostic radiologic examinations: collimation, exposure factors and use of protective equipment for the patients and their companions

INTRODUCTION

Producing appropriate diagnostic images along with patient radiation protection is the goal of radiography. Due to the advancements of radiography, concerns about observing the principles of radiation protection exist. Therefore, this study aimed to evaluate the observance of the principles of radiation protection in radiographic examinations with emphasis on field size collimation, suitability of exposure factors and the use of protective equipment for the patients and their companions.

METHODS

Using a cross-sectional study design, two radiography students on their final year of study observed 100 radiographic examinations from the imaging departments of five educational hospitals. The SPSS version 24 software was used to analyse the results.

RESULTS

The radiation field collimation was obtained in 46% of the studied radiographs. Patients had companions present during the examination in 26% of the studies; however, protective equipment was only used for 4% of the patients' companions, and no protective equipment was applied for patients. The observance rate of the various principles of radiation protection including field size restriction, the use of protective equipment for the patients and their companions, and suitability of the selected exposure factors was on average 44.6%.

CONCLUSION

The observance rate of the principles of radiation protection was insufficient in the studied educational hospitals, specifically in field size collimation and the use of protective equipment for the patients and their companions. Therefore, emphasis on the strict implementation of the radiation protection guidelines and continuous training of radiographers are required.

Introduction of a simple algorithm to create synthetic-Computed tomography of the head from magnetic resonance imaging

Background:

Recently, magnetic resonance imaging (MRI)-based radiotherapy has become a favorite science field for treatment planning purposes. In this study, a simple algorithm was introduced to create synthetic computed tomography (sCT) of the head from MRI.

Methods:

A simple atlas-based method was proposed to create sCT images based on the paired T1/T2-weighted MRI and bone/brain window CT. Dataset included 10 patients with glioblastoma multiforme and 10 patients with other brain tumors. To generate a sCT image, first each MR from dataset was registered to the target-MR, the resulting transformation was applied to the corresponding CT to create the set of deformed CTs. Then, deformed-CTs were fused to generate a single sCT image. The sCT images were compared with the real CT images using geometric measures (mean absolute error [MAE] and dice similarity coefficient of bone [DSC_bone]) and Hounsfield unit gamma-index (Г_HU) with criteria 100 HU/2 mm.

Results:

The evaluations carried out by MAE, DSC_bone, and Г_HU showed a good agreement between the synthetic and real CT images. The results represented the range of 78–93 HU and 0.80–0.89 for MAE and DSC_bone, respectively. The Г_HU also showed that approximately 91%–93% of pixels fulfilled the criteria 100 HU/2 mm for brain tumors.

Conclusion:

This method showed that MR sequence (T1w or T2w) should be selected depending on the type of tumor. In addition, the brain window synthetic CTs are in better agreement with real CT relative to bone window sCT images.

Calculation of Equivalent Square Fields in Radiation Therapy by an Improved Vadash Correction Factor for Collimator Exchange Effects

Background and objectives:

To determine the head scatter factor, a formalism presented by Vadash and Bjärngard has been employed to assess collimator exchange effects. The aim of this study was to determine the best Vadash correction factor (A) by introducing a new method based on the output factor measured in air for different square and rectangular fields.

Materials and Methods:

A new simple mathematical method based on selection of the best dosimetric data was proposed to obtain the A value for Vadash to predict the equivalent square field size. Measurements were performed with a Farmer chamber 0.6 cc in SSD 100 cm and build up cap of Plexiglas, 1.5 and 3.5 cm equal to water, for 6 and 18 MV, respectively, with a Perimus Plus linear accelerator. The output factor in air (OF_air) was measured for square and rectangular fields. MATLAB software (version R2014a) was employed for calculations and curve fitting.

Results:

A power model with a constant value was applied to the output factor in air as a function of square field size. The output factor in air ranged from 0.983 to 1.038 at 6MV and from 0.731 to 1.05 at 18 MV with the Y collimator having a greater effect. Obtained values for A were 1.42 and 1.55 with respect to σ_min=0.98 and 2.3 for 6 and 18 MV energies, respectively.

Conclusion:

The proposed A values minimize the collimator exchange effect in calculating equivalent squares, which plays an important role in patient dose calculation and treatment planning.

Analytical investigation of magnetic field effects on Proton lateral deflection and penetrating depth in the water phantom: A relativistic approach

Background

Integrated proton therapy - MRI systems are capable of delivering high doses to the target tissues near sensitive organs and achieve better therapeutic results; however, the applied magnetic field for imaging, influences the protons path, changes the penetration depth and deflects the particles, laterally, leading to dose distribution variations.

Objective

To determine the effects of a magnetic field on the range and the lateral deflection of protons, analytically.

Methods

An analytical survey based on protons energy and range power law relation, without using small angle assumption was done. The penetration depth and lateral deflection of protons with therapeutic energy ranges 60–250 MeV in the presence of uniform magnetic fields of 0–10T intensities, were calculated analytically. Calculations were done for relativistic conditions with Mathematica software version 7.0, and MATLAB 7.0 was applied to plot curves and curve fittings.

Results

In the presence of a magnetic field, the depth of Bragg peak was decreased and it was shifted laterally. A second order polynomial model with power equation for its coefficients and a power model with quadratic polynomial coefficients predicted the maximum lateral deflection (y_max) and maximum penetration depth (z_max) variations with energy and magnetic field intensity, respectively.

Conclusion

The applied correction for deflection angle will give more reliable results in initial energy of 250 MeV and 3T magnetic field intensity. For lower energies and magnetic field intensities the differences are negligible, clinically.

The association between infection burden in Iranian patients with acute myocardial infarction and unstable angina

AIM

to evaluate the association of ischemic heart disease (IHD) with the number of pathogens (infection burden) among individuals with infection.

METHODS

a total of 120 patients with IHD as the acute myocardial infarction (AMI; n=60) or unstable angina (UA; n=60) group and 60 healthy subjects with sex- and age-matched as control group were enrolled in this study. Serum samples of all participants were tested for the presence of antibodies to Helicobacter pylori (H. pylori), cytomegalovirus (CMV), type-1 herpes simplex virus (HSV-1) and type- 2 HSV (HSV-2) by using ELISA.

RESULTS

Regarding the association of the infection burden with IHD, the prevalence ratios and 95% confidence intervals (CI) were 3.18 (CI: 1.50-6.72; P<0.001) for 3 seropositivities and 3.83 (CI: 0.84-17.43; P<0.05) for 4 seropositivities. The rate of subjects with high infection burden (3 seropositivities) was significantly higher in IHD group as compared to control group (53.4% vs 21.6%; P<0.01). Moreover, the mean number of seropositivities was also significantly higher in patients with IHD in comparison to control group (2.47 vs 1.68; P<0.01). The seroprevalence of anti-H. pylori antibodies in AMI and UA groups was significantly higher compared to control group (P<0.0001). The seroprevalence of anti-CMV antibodies in AMI and UA group was also significantly higher than those observed in control group (P<0.01). Moreover, the seroprevalence of anti-HSV-1 antibodies was significantly higher in AMI and UA groups in comparison to control group (P<0.001). The seroprevalence of anti-HSV-2 antibodies was similarly expressed in patients and healthy control group.

CONCLUSION

the infection burden was significantly higher in patients with IHD, which represent that the parameter should also be considered as an independent risk factor for development of IHD. The seroprevalence of H. pylori, CMV and HSV-1 were also higher in patients with IHD.

Effect of adsorbents and chemical treatments on the removal of strontium from aqueous solutions

In the present investigation, three different solid wastes namely almond green hull, eggplant hull, and moss were initially treated and used as adsorbents for the adsorption of strontium ion from aqueous solutions. Adsorbent types and chemical treatments are proved to have effective roles on the adsorption of Sr(II) ion. Among the three adsorbents, almond green hull demonstrated strong affinity toward strontium ion in different solutions. The effectiveness of this new adsorbent was studied in batch adsorption mode under a variety of experimental conditions such as: different chemical treatments, various amounts of adsorbent, and initial metal-ion concentration. The optimum doses of adsorbent for the maximum Sr(II) adsorption were found to be 0.2 and 0.3 g for 45 and 102 mg L(-1) solutions, respectively. High Sr(II) adsorption efficiencies were achieved only in the first 3 min of adsorbent's contact time. The kinetics of Sr(II) adsorption on almond green hull was also examined and it was observed that it follows the pseudo second-order behavior. Both Langmuir and Freundlich models well predicted the experimental adsorption isotherm data. The maximum adsorption capacity on almond green hull was found to be 116.3 mg g(-1). The present study also confirmed that these low cost agriculture byproducts could be used as efficient adsorbents for the removal of strontium from wastewater streams.

Serum concentrations of Helicobacter pylori IgG and the virulence factor CagA in patients with ischaemic heart disease

To compare the serum concentrations of IgG to Helicobacter pylori and its virulence factor CagA in patients with ischaemic heart disease (IHD), we recruited 120 patients with IHD [acute myocardial infarction (AMI) (n = 60); unstable angina (UA) (n = 60)] and 60 sex- and age-matched healthy controls in this study. The seroprevalence of anti-H. pylori IgG was 86.7% in AMI, 91.7% in UA patients and 58.3% in the control group with mean titres of 33.2 U/ml [standard error (SE) 4.76], 57.96 U/ml (SE 7.54) and 25.72 U/ml (SE 4.01) respectively. The seroprevalence of anti-H. pylori in the patient groups was significantly higher than the control group. The mean levels of anti-H. pylori in the AMI and UA groups were also significantly higher than in the control group. The seroprevalence and mean titre of anti-CagA IgG did not differ significantly between patient and control groups.

Rapid removal of cobalt ion from aqueous solutions by almond green hull

Almond green hull, an agriculture solid waste, was chemically treated and used for the adsorption of Co (II) from aqueous solutions. The efficiency of this new adsorbent was studied using batch adsorption technique under different experimental conditions such as sorbent amount, initial metal-ion concentration, contact time, adsorbent particle size, and chemical treatment. Optimum dose of sorbent for maximum metal-ion adsorption were 0.25 g for 51.5 mg l(-1) and 0.4 g for 110 mg l(-1) solutions, respectively. High removal efficiencies of Co (II) were occurred in the first 1 min of sorbent contact time. The adsorption of Co (II) on almond green hull was also observed to follow the pseudo second-order kinetics. Adsorption isotherms were expressed by Langmuir and Freundlich adsorption models. The Langmuir adsorption model fits the experimental data reasonably well compared to the Freundlich model. The maximum adsorption capacity of this new sorbent was found to be 45.5 mg g(-1). The present study revealed that such a low cost material could be used as an efficient sorbent for the removal of cobalt from wastewater streams.

Water softening by combination of ultrasound and ion exchange

Ion exchange resin used in this work was styrene-divinylbenzene co-polymer with sulfonic acid group as a strong acid cation resin. This resin is particularly well suited for the removal of water hardness. In water treatment, commonly used softening processes are chemical precipitation and ion exchange. In this study, a combination of ultrasound and ion exchange was applied for reducing the hardness of water. The rate of exchange or kinetics of ion exchange is governed by several parameters. Therefore, important variables such as intensity of ultrasound, amount of resin, concentration of ions and contact time were investigated. The experimental data related to the removal of magnesium and calcium ions were fitted properly with Langmuir model. The kinetic of removal for both ions was pseudo-first-order. In point of mechanism, the internal porous and film diffusion were both effective in the process. The capacity of sorption and the velocity of removal were higher in the presence of ultrasound than control method and this is related to the cavitation process.