Machine learning can accurately predict pre-admission baseline hemoglobin and creatinine in intensive care patients.

Patients admitted to the intensive care unit frequently have anemia and impaired renal function, but often lack historical blood results to contextualize the acuteness of these findings. Using data available within two hours of ICU admission, we developed machine learning models that accurately (AUC 0.86-0.89) classify an individual patient's baseline hemoglobin and creatinine levels. Compared to assuming the baseline to be the same as the admission lab value, machine learning performed significantly better at classifying acute kidney injury regardless of initial creatinine value, and significantly better at predicting baseline hemoglobin value in patients with admission hemoglobin of <10 g/dl.

Transcalvarial brain herniation volume after decompressive craniectomy is the difference between two spherical caps.

Decompressive craniectomy (DC) is a surgical procedure used to relieve severely increased intracranial pressure (ICP) by removing a portion of the skull. Following DC, the brain expands through the skull defect created by DC, resulting in transcalvarial herniation (TCH). Traditionally, people measure only changes in the ICP but not in the intracranial volume (ICV), which is equivalent to the volume of TCH (V(TCH)), in patients undergoing DC. We constructed a simple model of the cerebral hemispheres, assuming the shape of the upper half of a sphere with a radius of 8 cm. We hypothesized that the herniated brain following DC also conforms to the shape of a spherical cap. Considering that a circular piece of the skull with a radius of a was removed, V(TCH) is the volume difference between 2 spherical caps at the operated side and the corresponding non-operated side, which represents the pre-DC volume underneath the removed skull due to the bilateral symmetry of the skull and the brain. Subsequently, we hypothesized that the maximal extent of TCH depends on a because of the biomechanical limitations imposed by the inelastic scalp. The maximum value of V(TCH) is 365.0 mL when a is 7.05 cm and the height difference between the spherical caps (Δh) at its maximum is 2.83 cm. To facilitate rapid calculation of V(TCH), we proposed a simplified estimation formula, Vˆ(TCH)=1/2A(2)Δh, where A=2a. With the a value ranging between 0 and 7 cm, the ratio between Vˆ(TCH) and V(TCH) ranges between 0.77 and 1.27, with different Δh values. For elliptical skull defects with base diameters of A and C, the formula changes to Vˆ(TCH)=1/2ACΔh. If our hypothesis is correct, surgeons can accurately calculate V(TCH) after DC. Furthermore, this can facilitate volumetric comparisons between the effects of DCs in skulls of varying sizes, allowing quantitative comparisons between ICVs in addition to ICPs.

PICO element detection in medical text without metadata: are first sentences enough?

Efficient identification of patient, intervention, comparison, and outcome (PICO) components in medical articles is helpful in evidence-based medicine. The purpose of this study is to clarify whether first sentences of these components are good enough to train naive Bayes classifiers for sentence-level PICO element detection. We extracted 19,854 structured abstracts of randomized controlled trials with any P/I/O label from PubMed for naive Bayes classifiers training. Performances of classifiers trained by first sentences of each section (CF) and those trained by all sentences (CA) were compared using all sentences by ten-fold cross-validation. The results measured by recall, precision, and F-measures show that there are no significant differences in performance between CF and CA for detection of O-element (F-measure=0.731±0.009 vs. 0.738±0.010, p=0.123). However, CA perform better for I-elements, in terms of recall (0.752±0.012 vs. 0.620±0.007, p<0.001) and F-measures (0.728±0.006 vs. 0.662±0.007, p<0.001). For P-elements, CF have higher precision (0.714±0.009 vs. 0.665±0.010, p<0.001), but lower recall (0.766±0.013 vs. 0.811±0.012, p<0.001). CF are not always better than CA in sentence-level PICO element detection. Their performance varies in detecting different elements.

Estimating postoperative skull defect volume from CT images using the ABC method.

OBJECTIVES: Surgeons often perform decompressive craniectomy to alleviate a medically-refractory increase of intracranial pressure. The frequency of this type of surgery is on the rise. The goal of this study is to develop a simple formula for clinicians to estimate the volume of the skull defect, based on postoperative computed tomography (CT) studies. METHODS: We collected thirty sets of postoperative CT images from patients undergoing craniectomy. We measured the skull defect volume by computer-assisted volumetric analysis (V(m)) and our own ABC technique (V(abc)). We then compared the volumes measured by these two methods. RESULTS: The V(m) ranged from 3.2 to 76.4 mL, with a mean of 38.9 mL. The V(abc) ranged from 3.8 to 71.5 mL, with a mean of 38.5 mL. The absolute differences between V(abc) and V(m) ranged from 0.05 to 17.5 mL (mean: 3.8±4.2). There was no statistically significant difference between V(abc) and V(m) (p=0.961). The correlation coefficient between V(abc) and V(m) was 0.969. In linear regression analysis, the slope was 1.00086 and the intercept was -0.0035 mL (r(2)=0.939). The residual was 5.7 mL. CONCLUSION: We confirmed that the ABC technique is a simple and accurate method for estimating skull defect volume, and we recommend routine application of this formula for all decompressive craniectomies.

Automatic measurement of midline shift on deformed brains using multiresolution binary level set method and Hough transform.

Midline shift (MLS) is an important quantitative feature clinicians use to evaluate the severity of brain compression by various pathologies. The midline consists of many anatomical structures including the septum pellucidum (SP), a thin membrane between the frontal horns (FH) of the lateral ventricles. We proposed a procedure that can measure MLS by recognizing the SP within the given CT study. The FH region is selected from all ventricular regions by expert rules and the multiresolution binary level set method. The SP is recognized using Hough transform, weighted by repeated morphological erosion. Our system is tested on images from 80 patients admitted to the neurosurgical intensive care unit. The results are evaluated by human experts. The mean difference between automatic and manual MLS measurements is 0.23 ± 0.52 mm. Our method is robust and can be applied in emergency and routine settings.

Automated assessment of midline shift in head injury patients.

OBJECTIVES: Midline shift (MLS) is an important quantitative feature for evaluating severity of brain compression by various pathologies, including traumatic intracranial hematomas. In this study, we sought to determine the accuracy and the prognostic value of our computer algorithm that automatically measures the MLS of the brain on computed tomography (CT) images in patients with head injury. PATIENTS AND METHODS: Modelling the deformed midline into three segments, we had designed an algorithm to estimate the MLS automatically. We retrospectively applied our algorithm to the initial CT images of 53 patients with head injury to determine the automated MLS (aMLS) and validated it against that measured by human (hMLS). Both measurements were separately used to predict the neurological outcome of the patients. RESULTS: The hMLS ranged from 0 to 30 mm. It was greater than 5 mm in images of 17 patients (32%). In 49 images (92%), the difference between hMLS and aMLS was <1 mm. To detect MLS >5 mm, our algorithm achieved sensitivity of 94% and specificity of 100%. For mortality prediction, aMLS was no worse than hMLS. CONCLUSION: In summary, automated MLS was accurate and predicted outcome as well as that measured manually. This approach might be useful in constructing a fully automated computer-assisted diagnosis system.