Journal of Medical Innovation and Technology

Urgent diagnosis and triage in Intracranial Haemorrages with Machine Learning

2021-01-09T19:13:57+00:00

Acute intracranial hemorrhages, regardless of their type, are pathologies with high mortality and require rapid diagnosis and treatment, however the patient group who will benefit most from early operation is operated later than the patient group with nless favorable outcome, because they do not admit with a severe clinical presentation. In this study, we aimed to evaluate a deep learning model that can distinguish the presence of intracranial hemorrhage in a small data set. Material Method: 3 healthy patients and 5 patients with intracranial hemorrhages were randomly seleceted for the study from the qure.ai Cranial CT database. The data set was created with a total of 200 CT cross-section images, 100 of which were hemorrhagic and 100 were healthy, and it was divided into three groups as training, validation and test set. The artificial neural network was trained in the training set and its accuracy was tested in the validation set, the accuracy did not improve after reaching around 80% and the training of the artificial neural network was stopped. Later, this artificial neural network was evaluated in the test set. Results: The deep learning model was run on the test set. Results were as follows; Sensitivity 90.0%, Specificity: 70.0%, Positive Predictive Value: 75.0%, Negative Predictive Value: 87.5% Total Accuracy: 80.0%. The deep learning model made only one false-negative assessment in 20 crosss-sections that it had never seen before. As a result, we think that a deep learning model can produce highly accurate results even if they are trained in a small data set and potentially be used for rapid triage in emergency departments.

Finite Element Analysis of Unicondylar Knee Arthroplasty Combined with Proximal Fibular Osteotomy

2021-01-09T19:14:00+00:00

Besides conservative methods, various surgical treatment modalities including high tibial osteotomy (HTO), unicondylar knee arthroplasty (UKA) and total knee arthroplasty (TKA) have been applied for the treatment of medial gonarthrosis. Combined application of UKA and PFO may enable longer implant survival by reducing the load over the implant and subsequent implant wear. We aimed to investigate on a finite element (FE) model, whether integrating a fibulectomy would reduce the contact stresses on the UKA implant. Finite Elements model developed to evaluate changes in the biomechanical environment resulting from UKA implantation and fibula osteotomy for use in the study. Menisci, bone and implants modeled as linear elastic isotropic material. We developed a 3D knee joint model with UKA implant and performed PFO on this model. A vertical load of 800N was applied anatomically to the femoral head. Stress on the tibial component was assessed before and after fibulectomy was performed. The contact stress was evaluated for UKA the mean stress (10,2 MPa ± 6,8) and for UKA + Fibulectomy (9,44 MPa ± 6,5). Combined application of PFO and UKA reduced contact stresses over the implant on a Finite Element Model. The findings of our observe additionally offer which might also additionally enhance the affected patient satisfaction, medical effects and implant survivorship of UKA.

The Mechanical Comparison of Artificial Bone and 3D Printed Bone Segments

2021-01-09T19:14:00+00:00

Bone is considered as an anisotropic structure due to the difference in its mechanical properties of cortical and spongiosal parts of the long bones. Researchers are attracted to bone related diseases and fractures in mechanical studies which leads them to seek alternative models. For decades artificial bones, especially Sawbones, are commonly preferred in biomechanical studies, which has similar density of natural bone. On the other hand, in the recent years there have been many studies by using 3D printer based bone models. In this study, we aimed to compare the artificial bone and 3D printed bone segments according to their mechanical properties. Cross sectional dimensions of an anatomical femur was examined with the Computed Tomography (CT) and a solid model was created by this data. Fused Deposit Manufacturing (FDM) technique and PLA filament was used in the specimen production. Two groups of bone segments produced by using a 3D printer at cortical thicknesses of 1.2 mm and 2.8 mm with a height of 10 mm. These groups were compared with sawbones cut in 10 mm heights. Biomechanical compression test was performed in three groups at a speed of 2 mm / min at 1000 N. As a result, the average of maximum force for sawbone, 1.2 mm and 2.8 mm thicknesses were 1006.3 N, 1009.5 N and 1010.6 N, respectively. Meanwhile, the average of maximum displacement for sawbone, 1.2 mm and 2.8 mm thicknesses were 0.203 mm, 0.183 mm and 0.191 mm, respectively. In conclusion, 3D printed bone models were found to be a good alternative for biomechanical analysis due to its similar force and displacement ratios.

Application of 3D Printers in Neurosurgery

2021-01-09T19:14:01+00:00

Utilisation of 3D printers has increased in medicine as well as other fields thanks to developing technology. There are lots of opportunities to utilize them in the medical field. There are several steps consisting image acquisition to surface post processing in 3D printing. In neurosurgery, there are lots of topics which can utilize 3D printing technology. Education of young neurosurgeons, preoperative planning and preperation, orientation to intraoperative anatomy, surgeon-patient communication and better understanding of informed consent period, design and production of new surgical instruments are among them. Besides, there are some drawbacks like, long duration of printing process, possible costs and resolution and texture quality of acquired models waiting for possible solutions. 3D printing is a promising technology in medical field. It has a potential to revolutionize surgical treatment as becomes widespread and cheaper.

T1 and T2 Mapping in Cardiac MRI

2021-01-09T19:14:02+00:00

In the recent years, according to the development in magnetic resonance imaging (MRI) , the parallel imaging techniques such as T1 and T2 mapping have been the guide for diagnosis. Mapping technics give ideas about tissue characterization. In this review, we mention about T1 and T2 mapping in cardiac MRI.

Radiomics Utilization in Neuro-Oncology: Brief Review of the State-of-art

2021-01-09T19:14:02+00:00

Radiomics is nascent field that involves extracting quantitative features - radiomic features from medical images that corrolate with the properties of the concerned lesion, such as the heterogeneity, shape, volume, proteomic, demographic and histology that have personalizied clinical value in diagnosis, prognosis and treatment. Our article aims to explore and review the state-of-art Radiomics and texture analysing in neuro-oncology, address the challenges a provide a prospect for future studies.

3D Printed New Surgical Tool For Complex Spinal Surgeries:A Technical Note

2021-01-09T19:14:02+00:00

3 dimensional (3D) printing technology is being used for preoperative evaluation of surgical anatomy and development of medical tools. This technology is thought to improve the surgical success by preoperative simulation of the surgical approach. Unique 3D printed models for every patient appears to decrease the peroperative complications. In this technical note, a novel angle meter tool obtained with 3D printing is proposed. The designing steps and use of this tool are desciribed. With the combined use of 3D printed anatomical model and angle meter, surgical orientation of the surgeon appears to be improved. There were no complications or malpositioned screws on the described patient. Therefore, together with the previous experiences of us on using 3D printed models intraoperatively, these models may be beneficial for decreasing screw malposition and operative duration.