Search published articles
Volume 56, Issue 4 (1 1998)
This paper represents a retrospective study of 142 patients with skeletal tuberculosis during the years 1350 to 1357 at Shafa Yahiaian Hospital. Confirmation of diagnosis is made by positive culture or pathology. The mean age of the patients was 27.5 years (range from 1.5 to 72 years). Most of the patients were in the second, first and third decade respectively. 45.3% of the patients were female. The spine was the most common site and was affected in 40.88% of the patients. The next sites were knee, hip and elbow, in order of frequency. The commonest site in spine was thoracic. The mean of sedimentation rate in the first visit was 48. Tuberculin skin test was negative in 23% of the patients. Evidence of pulmonary tuberculosis in chest radiography was present in 48.61% of the patients. Five patients had tuberculosis osteomylitis without involvement of the joint.
Volume 65, Issue 6 (3 2007)
Background: Depending on the lung volume in radiotherapy fields, breast cancer radiotherapy has documented side effects on pulmonary function, which can be determined by pulmonary function tests. Central lung distance (CLD), the distance from the chest wall to the edge of the field at the central axis, is an indicator of lung volume within the radiotherapy fields. In this study, we aim to detect the relationship between CLD and pulmonary function tests.
Methods: In this study we included 50 patients with breast cancer receiving postoperative adjuvant radiotherapy at Imam Hossein Hospital, Tehran, Iran. The patients received radiotherapy with a total dose of 4800-5000 cGy. For all patients, the central lung distances were measured using simulation of tangential fields, in addition to determination of pulmonary function, including force vital capacity (FVC) and forced expiratory volume in 1 s (FEV1) determined before radiotherapy, one month and three months after radiotherapy.
Results: There is no significant statistical difference between the FEV1 and FVC values before radiotherapy and those measured one month after radiotherapy however there was a significant statistical decrease in the FEV1 and FVC before radiotherapy and those measured three months after radiotherapy (P<0.001 and P<0.006, respectively). There is a positive statistical correlation between the change in the FEV1 three months after therapy and the CLD (r=0.71, p<0.01) and that of the FVC three months after therapy and the CLD (r=0.59, p<0.01). Linear regression for the prediction of FEV1 and FVC three months after radiotherapy was designed according to the CLD, FEV1 and FVC values before radiotherapy.
Conclusions: Three months after breast radiotherapy, the FEV1 and FVC values decrease, and the CLD is a proper predictor of these changes.
Volume 65, Issue 8 (3 2007)
Background: Congenital scoliosis is a developmental disorder defined as a lateral curvature of the spine. Its progressive trend and complications, such as cosmetic problems, pain and pulmonary symptoms, have put scoliosis as an important skeletal deformity that should be corrected. One of the currently accepted methods of treatment is posterior spinal fusion (PSF) that may be performed with or without instrumentation. However, the use of implants in conjunction with PSF in congenital spine deformity has been debated over the past three decades primarily because of increased risk of neurological deficit and implant displacement. The aim of this study was to compare short-term and long-term outcomes of spinal fusion with and without posterior instrumentation in congenital scoliosis.
Methods: In this historical cohort study, 41 patients with congenital scoliosis were recruited. All patients underwent PSF surgery between 1977 and 1996. They were divided into two groups according to the use of instrumentation: 22 congenital scoliotic patients who were treated by PSF without any instrumentation (group A), and 19 instrumented PSF patients (group B). Instrumentation was mostly performed using the Harrington rod. The major curve angle was measured before surgery, two weeks and one year after PSF surgery and at the end of the follow-up period.
Results: The mean baseline curve angles were 66.3° and 69.1° in groups A and B, respectively. The mean Cobb angles one year after PSF were 43.1° and 38.4° in groups A and B, respectively. The mean follow-up period was 8 years (SD=3) and, at the end of this period, the final Cobb angles were 47.3° and 39.4° in groups A and B, respectively. Therefore, the final angle correction was 28.7% in patients without instrumentation and 43% in patients with instrumentation. The mean loss of correction was 5.5% and 4.3% in groups A and B, respectively. The final curve angles was significantly more corrected for those patients in whom instrumentation was used than those without instrumentation (p<0.05).
Conclusion: The treatment of congenital scoliosis can be very challenging despite the benefits of modern surgeries and instrumentation methods. The results of our study demonstrate that the application of an implant with PSF surgery may lead to increased improvement of the scoliotic curvature in the short-term and long-term periods, as well as a decrease in the loss of correction and the rate of reoperation. In addition, the low incidence of complications in our study indicates the safety of the posterior instrumentation for the treatment of congenital scoliosis.
Volume 66, Issue 11 (3 2009)
Background: Degenerative spondylolisthesis is a common disease of the lumbar spine especially in older ones. The disease represents a challenge to the treating physician. At present, for those patients that deteriorate clinically, there are many proposed algorithms for the surgical treatment. This before and after study was undertaken to assess the surgical results of decompression and instrumented posterolateral fusion in these patients.
Methods: The study population consisted of 23 patients who had undergone no prior surgery for degenerative spondylolisthesis on the lumbar spine. These patients were treated by decompression, bilateral posterolateral fusion, and segmental (pedicle screw) instrumentation with mean follow-up of 29 months (range, 13-73 months). Finally, The clinical results were evaluated for all patients by means of an Oswestry Disability Index (ODI) version 2.1, the Henderson's functional capacity, and persistence of leg symptoms, low back pain or claudication. Mann-Whitney and Chi-Square tests were used to assess the average values and comparison, respectively.
Results: Henderson's functional capacity at the last visit session was excellent in 14 (60.9%), good in 7 (30.4%), fair in 2 (8.7%) cases. ODI decreased from 72.2% (50-88%) preoperatively to 14.4% (0-54%) at the latest follow-up visit. A history of leg pain or claudication was correlated significantly with the amount of decline in ODI score and Henderson's functional capacity (p<0.05).
Conclusion: In spite of limited number of our patients, decompressive surgery plus instrumented posterolateral fusion is a safe, reliable, and satisfactory procedure for treating degenerative lumbar spondylolisthesis. This procedure may be done when conservative treatment was failed and psychological problems can be ruled out.
Volume 69, Issue 5 (6 2011)
Background: Scoliosis is one of the most common spinal deformities with subsequent decrease in pulmonary function. The effects of surgical correction on the pulmonary function of patients with adolescent idiopathic scoliosis are controversial. The purpose of the present study was to compare the postoperative pulmonary function changes in different surgical approaches chosen for its correction.
Methods: Sixty-five patients with adolescent idiopathic scoliosis who had undergone corrective spinal surgery in Shafa Yahyaian Hospital since 1997 to 2007 and had documented preoperative pulmonary function test report, included in our study. The patients had documented preoperative pulmonary function tests and were divided into three groups based on their surgical approach. The first group was comprised of 25 patients who had undergone posterior spinal fusion (PSF), the second group included 29 patients who had anterior and posterior spinal fusion (ASF+PSF), and the third group consisted of 10 patients who had posterior spinal fusion and thoracoplasty (PSF+ thoracoplasty). The preoperative and the final follow-up values of absolute and percent predicted forced vital capacity (FVC) and forced expiratory volume in one second (FEV1) were compared and their relations with the undertaken surgical approach were evaluated.
Results: There were no significant correlations between the surgical approaches and changes in the pulmonary function tests (P>0.05). FVC and FEV1 values were similar to preoperative values in all groups after a minimum follow-up of two years.
Conclusion: There were no significant differences between different surgical approaches for adolescent idiopathic scoliosis regarding the pulmonary function test results after at least two years of follow-up.
Volume 70, Issue 8 (5 2012)
Background: Several studies have suggested higher incidence of osteoporosis in patients with idiopathic scoliosis in comparison with the normal population. The aim of this study was to assess the prevalence of low bone mass among adolescent girls with idiopathic scoliosis.
Methods: In this cross-sectional study performed in shafa Hospital in Tehran, Iran during 2011-2012, we recruited fifty-seven 12- to-20-year old girls with idiopathic scoliosis and compared them with 100 age-matched healthy girls. The patients had no other diseases including neuromuscular disorders, congenital vertebral anomalies or a history of spinal surgery. Bone mineral densities (BMD) of the hip and spine were evaluated and compared in all 157 participants using dual X-ray absorptiometry (DXA). Standard BMD (sBMD) was also calculated at the lumbar spine.
Results: Analysis of the data revealed that hip BMD was significantly (P=0.004) lower in patients with idiopathic scoliosis versus the controls. Moreover, BMD and sBMD of the Spine were also significantly lower in the patients (respectively, P=0.030 and P=0.030). Curve location had no effect on the values of hip BMD, spine BMD or spine sBMD (respectively, P=0.061 and P=0.274 and P=0.208).
Finally, with more severe curves a lower bone mass was detected for sBMD and spine BMD (respectively, P=0.017 and P=0.016), but it was not significant for hip BMD (P=0.069).
Conclusion: Adolescent girls with idiopathic scoliosis had lower bone mass compared with their healthy peers. The lower bone mass was correlated with the severity of the curve but not its location.
Volume 71, Issue 7 (October 2013)
Background: Damages caused by scald burns are common and can cause severe complications and death. The purpose of this study was to define risk groups and then methods of prevention and treatment is designed to fit.
Methods: Data for this retrospective study of hospitalized patients in Shahid Motahari Hospital in Tehran from 2007-2011 were compiled. Data including age, sex, cause of burn, and degree of burn and ultimate fate of the victims were collected from scald burns. Burns caused by boiling water and hot food (Scald) , in two age groups : 12 and under 12 years ( children) and more than 12 years ( adults) were compared in terms of statistics .
Results: A total of 1150 patients consisting of males (57.9%) and females (42.1%) were studied. The most common age was 1 year old and 50% of patients were under 3 years of age. 87.9% burned with boiling water and 12.1% had experienced burns with hot food. Incentive to burn was 0.3% cross burning and 99.7% incident. A maximum number of burns in children 12 years and younger males (42.1%) and a minimum number in men over 12 years (15.7%) were observed. Mean percentage of burns was 11% in over 12 years group and 30.9% in 12 and under 12 years group. The average hospital stay was 11.4 days and the mortality rate was 4.8%. The final status of the patients was as fallows: full recovery 904 cases (78.6%), partial recovery 134 (11.7%), clearance with personal consent 41 (3.6%), death 55 (4.8%) and 16 cases (3.1%) were among other reasons.
Conclusion: In general it can be said, scald burns incidence in individuals aged 12 and younger were more than the older ones and the mean of burns was lower in individuals with over 12 years old. There was a sexual preference for males under 12 years. Mortality rate in the two groups has not any statistically significant difference. There was no statistically significant association between sex and mortality rate. Some of our findings are depending on cultural, social and economic conditions, so generalized this findings to other geographical areas should be done with caution.
Volume 72, Issue 3 (June 2014)
Background: The rate of using firearm in criminal acts, even in countries with strict le-gal controls on these weapons is increasing. The most common methods of homicides and suicides are using firearm in United States. In this study we aimed to evaluate the epidemiology of gunshot bodies referred to Forensic Medicine. Methods: Information about firearm-related injuries in autopsy hall of Forensic Medi-cine center in Tehran during 2011-2012 were evaluated from patients’ documents. Necessary information including age, sex, education, employment, drug and alcohol addiction, as well as information related to shot such as suicide or homicidal, shoot number (incoming bullets), shot location, type of firearm, organ damage, shot distance and location of the incident were extracted and analyzed using SPSS version 16. Results: In this cross sectional study, 38 deaths were identified. These were over-whelmingly males (100%) with mean age of 32.4 (SD= 6.57). The typical entrance wound sites in suicides were the head (42.2%) and chest (24.3%) and abdomen (5.4%) and limb (2.7%) and others (48.6%). place a bullet in the head, 16 (42.2 %), chest 9 (24.3%), the abdomen, 2 cases (5.4%), an organ (2.7%) and hit a different place (9 cas-es 24.3 percent). Brain damaged in 18 cases (84.6%) and lung 5 (13.5%) and 12 (32.4%) and several organs were damaged. The typical entrance wound sites in suicides were the head (42.2%) and chest (24.3%) and abdomen (5.4%) and limb (2.7%) and others (48.6%). This total included 33% attempted or completed suicides, and 47.5% assaults/homicides The scene of disaster were the 12 patients (50%) outdoors, 8 (33.3%) at home and in 4 cases (16.7%) at work. Conclusion: Based on results of this study, age of victims is higher than previously ob-served patterns, such as gender of victims, location of shots, type of weapon used is different with international reports due to differences in availability of Iranian weapons and cultural differences. The results of this study can be a base for other investigations’s changes in trends of total firearm death rates, mass fatal shooting incidents, rates of firearm homicide, suicide and unintentional firearm deaths, and of total homicides and suicides.
Volume 73, Issue 1 (April 2015)
Background: Orthopedic trauma is a common type of injury in children and may cause deep and permanent psychological and physical damage both for the patient and the parents. This study aimed to analyze the epidemiology age, gender distribution and the mechanism of injury in patients presenting to a level I trauma center in urban population of Tehran. Methods: In this prospective descriptive study, the patients under 19 years old with orthopedic trauma who were hospitalized in Tehran Shafa University Hospital were entered. This hospital is the main orthopedic referral center in Iran. The patients were prospectively evaluated from April 2013 to March 2014. The data were collected and analyzed. Results: The study included 1081 patients under 19 years old. There was a male predominance (76.8% n= 830). The boys had a higher mean age 11.04±5.06 year, versus girls with mean age 8.67±4.63 year (P< 0.05). The peak age of boys was 18 and the girls had two peaks at three and nine. The fractures occurred in upper limb in 70.8% (n= 621) and 29.2% (n= 256) in lower limb of patients. There were 27 cases with joint dislocation, 5 cases with knee ligamentous injuries, 128 cases with soft tissue injuries and 44 cases with spine injuries. The most frequent mechanism in both gender were falls from standing position (48.5%). The most common fractures were foreman both bone fractures (n: 146 16.7%), elbow supracondylar fractures (n: 134, 15.3%) and distal radius fractures (n: 84, 9.6%). The most fractures occurred in summer (30.1%) and the least in winter (18.1%). Conclusion: Evaluation of epidemiologic factors can lead to the best prediction and treatment planning of trauma. Early recognition of injury, even minor, and expected care using specialized teams will help to improve outcomes for these patients. This study determines the most at risk children for trauma and fractures and may help the parents to prevent damage.
Volume 73, Issue 10 (January 2016)
Background: Acute respiratory distress syndrome (ARDS) is one of the most important complications associated with traumatic brain injury (TBI). ARDS is caused by inflammation of the lungs and hypoxic damage with lung physiology abnormalities associated with acute respiratory distress syndrome. Aim of this study is to determine the epidemiology of ARDS and the prevalence of risk factors.
Methods: This prospective study performed on patients with acute traumatic head injury hospitalization in the intensive care unit of the Shohaday-e Haftom-e-Tir Hospital (September 2012 to September 2013) done. About 12 months, the data were evaluated. Information including age, sex, education, employment, drug and alcohol addiction, were collected and analyzed. The inclusion criteria were head traumatic patients and exclusion was the patients with chest trauma. Questionnaire was designed with doctors supervision of neurosurgery. Then the collected data were analysis.
Results: In this study, the incidence of ARDS was 23.8% and prevalence of metabolic acidosis was 31.4%. Most injury with metabolic acidosis was Subarachnoid hemorrhage (SAH) 48 (60%) and Subdural hemorrhage (SDH) was Next Level with 39 (48%) Correlation between Glasgow Coma Scale (GCS) and Respiratory Distress Syndrome (ARDS) were significantly decreased (P< 0.0001). The level of consciousness in patients with skull fractures significantly lower than those without fractures (P= 0.009) [(2.3±4.6) vs (4.02±7.07)]. Prevalence of metabolic acidosis during hospitalization was 80 patients (31.4%).
Conclusion: Acute respiratory distress syndrome is a common complication of traumatic brain injury. Management and treatment is essential to reduce the mortality. In this study it was found the age of patients with ARDS was higher than patients without complications. ARDS risk factor for high blood pressure was higher in men. Most victims were pedestrians. The most common injury associated with ARDS was SDH. Our analysis demonstrates that Acute respiratory distress syndrome is common after traumatic brain injury. Management of traumatic brain injury is necessary to manage and reduce the mortality.
Volume 75, Issue 8 (November 2017)
Methods: The study group comprised of a random sample of 100 healthy people without congenital or acquired skeletal malformation or other obvious diagnosed disease, with the age range from 19 to 35 years in Tehran and Kashan cities, Iran, from March to the end of August of 2017. The number of male and female samples were equal. The lengths of second (2D), third (3D), forth (4D) and fifth (5D) finger of both hands were measured from the basal crease of the finger proximal to the palm to the tip of the finger using a ruler with a precision of 0.01 millimeters. Exploratory analysis were performed. Different finger ratios including 2D:3D, 2D:4D, 2D:5D, 3D:4D, 3D:5D and 4D:5D were calculated.
Results: The mean ratio of right fingers R 2:3, 2:4, 2:5, 3:4, 3:5, 4:5 and mean ratio of left fingers L 2:3, 2:4, 2:5, 3:4, 3:5, 4:5 were not equal in the groups compared. The index and ring finger ratio is found to be higher in males than females. Ratios of L 2:4, 3:4 and R 2:3, 2:4, 2:5 showed higher values in males.
Conclusion: Our results suggest that the 2D:4D ratio is the most decisive ratio which can demarcate between male and female in Iranian population. It may prove useful to determine the sex of an isolated hand in medicolegal examinations.
Volume 76, Issue 6 (September 2018)
Volume 76, Issue 8 (November 2018)
Methods: This prospective study was conducted on the X-ray images of adult people hands between the months of March and September 2017. They referred to Shafayahyaian Hospital, the educational and referral center of orthopedic disease of Iran University of Medical Sciences, in Iran. 200 adult persons (100 male and 100 female) without any background of specific disease, entered in our study. In each case, the length of D: distal phalanx, M: middle phalanx, P: proximal phalanx, MC: metacarpal, Fl: finger length, and T: total (Distal phalanx+middle phalanx+proximal phalanx+soft tissue of the finger tip) were measured in millimeter and reported in ratio. All analyses were done using SPSS software, version 20 (SPSS Inc., Chicago, IL, USA).
Results: Student t-test showed that many ratios were significantly different between two genders. It is concluded that the length ratio of metacarpals and phalanges could be used for sex determination. The most valuable ratios were FL1/FL3 and MC4/MC5. Multivariate logistic regression test reported an equation by which the gender is predictable. Z= 5.856+ -3.904 X FL1/FL3+ -1.865 X MC4/MC5.
Conclusion: If “Z=0”, the gender is male and if “Z=1”, it is female without any doubt. In our knowledge, this study provides the first population sex estimation formulae from the hand bones in Iranian population. However, we need further studies in larger groups to use this equation for sex determination in forensic contexts.
Volume 77, Issue 2 (May 2019)
Methods: This cross-sectional study was carried out at the Radiography Centers of Semnan University of Medical Sciences, Iran, from July 2017 to September 2017. First, patients who were under radiography, were asked about the knowledge of the radiographers about the 10-day rule and radiation protection. Then a questionnaire with 12 questions was given to the radiographers in radiology and CT scan center. Finally, the existence of protection guidelines for pregnant women was determined by asking manager and checking at the moment.
Results: The patient's question about pregnancy failed about 19%, which did not have a significant relationship with patients' singleness (P=0.0004). Also the mean scores for the knowledge of radiographers about radiation protection were 14.21±0.96, which did not have a significant relationship with their work record and place of employment and their educational level (P=0.09). On the other hand, in half of the radiography centers, the written radiation protection guidelines were not available.
Conclusion: Principles of radiation protection and 10-day rule for radiography of women who were prone to pregnancy were desirable but due to importance of radiation protection, the rules should be more carefully implemented.
Volume 77, Issue 7 (October 2019)
Methods: A myoelectric system based on convolutional neural networks (CNN) is proposed, as an alternative to conventional classification methods that depend on feature engineering. The proposed model was validated with 10 able-bodied subjects during single and combined wrist motions. Eight EMG channels were recorded using eight pairs of surface electrodes attached around the subject’s dominant forearm. The raw EMG data from windows of 167ms (200 samples) in 8 channels were arranged as 200×8 matrices. For each subject, a CNN was trained using the EMG matrices as the input and the corresponding motion classes as the target. The resulting model was tested using a 4-fold cross-validation. The performance of the proposed approach was compared to that of a standard SVM-based model that used a set of time-domain (TD) features including mean absolute value, zero crossings, slope sign changes, waveform length, and mean frequency.
Results: In spite of the proven performance and popularity of the TD features, no significant difference (P=0.19) was found between the classification accuracies of the two methods. The advantage of the proposed model is that it does not need manual extraction of features, as the CNN can automatically learn and extract required representations from the EMG data.
Conclusion: These results indicate the capacity of CNNs to learn and extract rich and complex information from biological signals. Because both amplitude and frequency of EMG increases with increasing muscle force, both temporal and spectral characteristics of EMG are needed for efficient estimation of motor intent. The TD set, also includes these types of features. The high performance of the CNN model shows its capability to learn temporal and spectral representations from raw EMG data.
Volume 77, Issue 9 (December 2019)
Case Presentation: The patient was an unmarried 32-year-old man, who was hospitalized in Rasul-Akram Hospital in December 2018 for severe bleeding caused by GSM. He was diagnosed with type I bipolar disorder, which was treated from the late adolescence. Also, a history of drug abuse (amphetamine, opium, cannabis, alcohol) and multiple sexual relationships with different partners was reported. He abused tramadol tablets daily nowadays. His IQ was normal. He had lots of abrasions, scars and lacerations due to suicidal attempts in the past years. Two months before GSM, his auditory hallucinations about genital self-mutilation had started. He committed GSM after having sexual intercourse and was transferred to our hospital by an ambulance with severe hemorrhage, scrotum damage, and presentation of both testes, for which he received appropriate treatments.
Conclusion: It seems that a set of factors, such as patient's history of psychosis following bipolar disorder, substance abuse, mental retardation, multiple sexual relationships, and strong emotional relationship with the mother can contribute to self-injury that may lead to genital self-mutilation. It should be noted that genital self-mutilation cases are less likely to be reported in Iran, considering the cultural and religious background. Therefore, the pathology of this devastating phenomenon should be more thoroughly examined.
Volume 78, Issue 3 (June 2020)
Methods: In this analytic study, the database of HAM10000 (human against machine with 10000 training images) dermoscopy images, 1000 melanocytic nevi and 1000 melanoma images were employed, where in each category 900 images were selected randomly and were designated as the training set. The remaining 100 images in each category were considered as the test set. A deep learning convolutional neural network (CNN) was deployed with AlexNet (Krizhevsky et al., 2012) as a pretrained model. The network was trained with 1800 dermoscope images and subsequently was validated with 200 test images. The proposed method removes the need for cumbersome tasks of lesion segmentation and feature extraction. Instead, the CNN can automatically learn and extract useful features from the raw images. Therefore, no image preprocessing is required. Study was conducted at Shahid Beheshti University of Medical Sciences, Tehran, Iran from January to February, 2020.
Results: The proposed model achieved an area under the receiver operating characteristic (ROC) curve of 0.98. Using a confidence score threshold of 0.5, a classification accuracy of 93%, sensitivity of 94%, and specificity of 92% was attained. The user can adjust the threshold to change the model performance according to preference. For example, if sensitivity is the main concern; i.e. false negative is to be avoided, then the threshold must be reduced to improve sensitivity at the cost of specificity. The ROC curve shows that to achieve sensitivity of 100%, specificity is decreased to 83%.
Conclusion: The results show the strength of convolutional neural networks in melanoma detection in dermoscopy images. The proposed method can be deployed to help dermatologists in identifying melanoma. It can also be implemented for self diagnosis of photographs taken from skin lesions. This may facilitate early detection of melanoma, and hence substantially reduce the mortality chance of this dangerous malignancy.
Volume 78, Issue 4 (July 2020)
Methods: In this analytic study, 327 AKIEC, 513 BCC, and 840 benign keratosis images from human against machine with 10000 training dermoscopy images (HAM10000) were extracted. From each of these three types, 90% of the images were designated as the training set and the remaining images were considered as the test set. A deep learning convolutional neural network (CNN) was developed for skin cancer detection by using AlexNet (Krizhevsky, et al., 2012) as a pretrained network. First, the model was trained on the training images to discriminate between benign and malignant lesions. In comparison with conventional methods, the main advantage of the proposed approach is that it does not need cumbersome and time-consuming procedures of lesion segmentation and feature extraction. This is because CNNs have the capability of learning useful features from the raw images. Once the system was trained, it was validated with test data to assess the performance. Study was carried out at Shahid Beheshti University of Medical Sciences, Tehran, Iran, in January and February, 2020.
|
Results: The proposed deep learning network achieved an AUC (area under the ROC curve) of 0.97. Using a confidence score threshold of 0.5, a classification accuracy of 90% was attained in the classification of images into malignant and benign lesions. Moreover, a sensitivity of 94% and specificity of 86% were obtained. It should be noted that the user can change the threshold to adjust the model performance based on preference. For example, reducing the threshold increase sensitivity while decreasing specificity.
Conclusion: The results highlight the efficacy of deep learning models in detecting non-melanoma skin cancer. This approach can be employed in computer-aided detection systems to assist dermatologists in identification of malignant lesions. |
Volume 79, Issue 5 (August 2021)
Breast cancer is one of the most common types of cancer in women. Screening mammography is a low‑dose X‑ray examination of breasts, which is conducted to detect breast cancer at early stages when the cancerous tumor is too small to be felt as a lump. Screening mammography is conducted for women with no symptoms of breast cancer, for early detection of cancer when the cancer is most treatable and consequently greatly reduce the death rate from the breast cancer. Screening mammography should be performed every year for women age 45-54, and every two years for women age 55 and older who are in good health. A mammogram is read by a radiologist to diagnose cancer.
To assist radiologists in reading mammograms, computer-aided detection (CAD) systems have been developed which can identify suspicious lesions on mammograms. CADs can improve the accuracy and confidence level of radiologists in decision making and have been approved by FDA for clinical use. Traditional CAD systems work based on conventional machine learning (ML) and image processing algorithms. With recent advances in software and hardware resources, a great breakthrough in deep learning (DL) algorithms was followed, which revolutionized various engineering areas including medical technologies. Recently, DL models have been applied in CAD systems in mammograms and achieved outstanding performance. In contrast to conventional ML, DL algorithms eliminate the need for the tedious task of human-designed feature engineering, as they are capable of learning useful features automatically from the raw data (mammogram). One of the most common DL frameworks is the convolutional neural network (CNN). To localize lesions in a mammogram, a CNN should be applied in region‑based algorithms such as R‑CNN, Fast R‑CNN, Faster R‑CNN, and YOLO.
| Proper training of a DL‑based CAD requires a large amount of annotated mammogram data, where cancerous lesions have been marked by an experienced radiologist. This highlights the importance of establishing a large, annotated mammogram dataset for the development of a reliable CAD system. This article provides a brief review of the state‑of‑the‑art techniques for DL‑based CAD in mammography. |
| Page 1 from 1 |
Related Websites
Site Keywords
Cardiology, Anatomy, Biology, Epidemiology, Immunology, Neuroscience, Physical therapy, Physiology, Dermatology, Endocrinology and metabolicSite Statistics
- Registered users: 7661 users
- Online users: 0 users
- Guest users: 261 users
- All visits: 14532005 visits
- Visits in 24 Hours: 6285 visits
- Total articles: 11113 articles
- Published articles: 4890 articles
© 2024 , Tehran University of Medical Sciences, CC BY-NC 4.0
Designed & Developed by : Yektaweb