Deep Learning and Multimodal Artificial Intelligence in Orthopaedic Surgery.

This review article focuses on the applications of deep learning with neural networks and multimodal neural networks in the orthopaedic domain. By providing practical examples of how artificial intelligence (AI) is being applied successfully in orthopaedic surgery, particularly in the realm of imaging data sets and the integration of clinical data, this study aims to provide orthopaedic surgeons with the necessary tools to not only evaluate existing literature but also to consider AI's potential in their own clinical or research pursuits. We first review standard deep neural networks which can analyze numerical clinical variables, then describe convolutional neural networks which can analyze image data, and then introduce multimodal AI models which analyze various types of different data. Then, we contrast these deep learning techniques with related but more limited techniques such as radiomics, describe how to interpret deep learning studies, and how to initiate such studies at your institution. Ultimately, by empowering orthopaedic surgeons with the knowledge and know-how of deep learning, this review aspires to facilitate the translation of research into clinical practice, thereby enhancing the efficacy and precision of real-world orthopaedic care for patients.

Extreme weekly locoregional hypofractionated radiation in elderly women with non-metastatic breast cancer.

BACKGROUND AND PURPOSE: Breast cancer locoregional (LR) radiation in the elderly requires careful consideration between the benefits of aggressive treatment and its potential toll on these patients. Extreme weekly LR hypofractionated radiation (HFRT), delivering >5 Gy per fraction, may be better suited in such a population. It represents a good compromise between RT omission and exhaustive daily radiation. This study aims to report the local and LR control rate as well as the acute and long-term side effects of the elderly patients treated with HFRT in our institution, and to compare these results to those from the literature. MATERIALS AND METHODS: We conducted a retrospective study by reviewing medical records of elderly patients with breast cancer treated with adjuvant once-weekly LR HFRT between 2011 and 2020. Fifty patients presenting with primary non-metastatic node-positive breast tumors were included. Treatment outcomes including local/LR control and overall survival were reported. Early and late toxicity profiles were also assessed. RESULTS: After a median follow-up of 4.8 years, only one local recurrence in the chest wall occurred and there was no regional recurrence. The distant metastatic rate was 6%. The long-term recurrence-free survival rate was 80% at 5 years. The cause specific survival rate was 90% at 5 years. The overall survival rate was 69.4% and 55.5% at 3 and 5 years, respectively. There were 44 (88%) patients with Grade 1 or 2 early toxicity. There was no Grade 3 or higher acute toxicity registered. Late toxicity was mainly Grade 1 or 2 subcutaneous fibrosis, lymphoedema, and neuropathy except for one patient with Grade 3 fibrosis. CONCLUSION: Extreme LR HFRT is well tolerated with good outcomes and is a good alternative for elderly and frail patients. Our results confirm the efficacy and safety of such a regimen. Further randomized trials assessing both oncologic outcome and toxicity profile are justified.

The Clinical Significance of Bone Mineral Density Changes Following Long-Term Androgen Deprivation Therapy in Localized Prostate Cancer Patients.

PURPOSE: Long-term androgen deprivation therapy has been associated with decreased bone mineral density in men with prostate cancer. Some evidence suggests that there is no impact on fracture risk despite this bone mineral density loss. Our study aimed to quantify changes in bone mineral density in men with high risk prostate cancer on long-term androgen deprivation therapy and calcium and vitamin D supplementation. MATERIALS AND METHODS: Bone mineral density analysis was conducted for localized high risk prostate cancer patients enrolled in the phase III randomized trial PCS-V (Prostate Cancer Study 5), comparing conventional and hypofractionated radiation therapy. Patients received 28 months of luteinizing hormone-releasing hormone agonist and calcium and vitamin D supplementation (500 mg calcium BID+400 IU vitamin D3 BID). The areal density and T-scores (spine, femoral neck and total femur) at baseline and 30 months of followup were extracted, and the absolute change was calculated. Clinical bone density status (normal, osteopenia, osteoporosis) was monitored. RESULTS: The lumbar spine, femoral neck and total femoral bone mineral density were measured for 226, 231, and 173 patients, respectively. The mean percent change in bone mineral density was -2.65%, -2.76% and -4.27% for these respective sites (p <0.001 for all). The average decrease in bone mineral density across all sites was -3.2%, with no decline in bone mineral density category in most patients (83%). Eight patients (4%) became osteoporotic. CONCLUSIONS: Despite a mild decline in bone mineral density, the change in clinical bone mineral density category remained low with long-term androgen deprivation therapy. Consequently, calcium and vitamin D supplementation alone may suffice for most localized prostate cancer patients on long-term androgen deprivation therapy.

Combined Long-Term Androgen Deprivation and Pelvic Radiotherapy in the Post-operative Management of Pathologically Defined High-Risk Prostate Cancer Patients: Results of the Prospective Phase II McGill 0913 Study.

Purpose: Following radical prostatectomy, prostate bed radiotherapy (PBRT) has been combined with either long-term androgen deprivation therapy (LT-ADT) or short-term ADT with pelvic lymph node radiotherapy (PLNRT) to provide an oncological benefit in randomized trials. McGill 0913 was designed to characterize the efficacy of combining PBRT, PLNRT, and LT-ADT. It is the first study to do so prospectively. Methods: In a single arm phase II trial conduced from 2010 to 2016, 46 post-prostatectomy prostate cancer patients at a high-risk for relapse (pathological Gleason 8+ or T3) were assessed for treatment with combined LT-ADT (24 months), PBRT, and PLNRT. Patients received PLNRT and PBRT (44 Gy in 22 fractions) followed by a PBRT boost (22 Gy in 11 fractions). The primary endpoint was progression-free survival (PFS). Toxicity and quality of life (QoL) were evaluated using CTCAE V3.0 and EQ-5D-3L questionnaires, respectively. Results: Among the 43 patients were treated as per protocol, median PSA was 0.30 µg/L. On surgical pathology, 51% had positive margins, 40% had Gleason 8+ disease, 42% had seminal vesicle involvement, and 19% had lymph node involvement. At a median follow-up of 5.2 years, there were no deaths or clinical progression. At 5 years, PFS was 78.0% (95% Confidence Interval 63.7-95.5%). Not including erectile dysfunction, patients experienced: 14% grade 2 endocrine toxicity while on ADT, one incident of long-term gynecomastia, 5% grade 2 acute urinary toxicity, 5% grade 2 late Urinary toxicity, and 24% long-term hypogonadism. No comparison between the average or minimum self-reported QoL at baseline, during ADT, nor after ADT demonstrated a statistically significant difference. Conclusions: Combining PBRT, PLNRT, and LT-ADT had an acceptable PFS in patients with significant post-operative risk factors for recurrence. While therapy was well-tolerated, long-term hypogonadism was a substantial risk. Further investigations are needed to determine if this combination is beneficial. Trial registration: NCT01255891.

Tumor Board Shadowing: A Unique Approach for Integrating Radiation Oncologists Into General Medical Student Education.

PURPOSE: Radiation oncology is often overlooked in US medical school curricula, with few opportunities for most students to learn about the specialty or the value of radiation therapy in cancer care. Tumor boards represent a potential avenue not only to increase students' exposure to radiation oncologists but also to provide a fundamental understanding of the multidisciplinary nature of cancer care and effective collaboration in clinical practice. METHODS AND MATERIALS: In this study, we evaluated a novel radiation oncologist-driven tumor board shadowing experience at 3 medical schools in the United States and Canada. A total of 323 first- and second-year medical students participated, of whom 77.4% completed a follow-up survey assessing the effectiveness of the program as a learning tool. RESULTS: Compared with traditional clinical shadowing, students were more likely to believe that tumor board shadowing provided a similar or better experience in terms of educational content (85%), exposure to a new field (96%), and overall experience (89%). Forty-eight percent of students perceived a greater amount of multidisciplinary collaboration in oncologic care than they thought existed prior to attending. Forty-eight percent of students also felt more competent interacting with oncologists after participating, whereas 21% felt more competent interacting with patients with cancer. Students' perception of increased competence was correlated with the amount of time their assigned physician mentor spent answering their questions after the tumor board (P < .01). Second-year medical students also had a more favorable overall experience than first-year medical students did (P = .04). CONCLUSIONS: Multidisciplinary tumor boards can be used effectively as a unique immersive learning opportunity that can be feasibly implemented to improve knowledge of clinical oncology and multidisciplinary care in medical schools and expose students to physicians in smaller fields such as radiation oncology.

Does Interfraction Cone Beam Computed Tomography Improve Target Localization in Prostate Bed Radiotherapy?

PURPOSE: In this prospective phase II study, we investigated whether cone beam computed tomography scan was a superior method of image-guided radiotherapy relative to 2D orthogonal kilovoltage images in the post-radical prostatectomy setting. METHODS: A total of 419 treatment fractions were included in this analysis. The shifts required to align the patient for each treatment were performed using 3D matching between cone beam computed tomography scans and the corresponding computed tomography images used for planning. This was compared with the shifts obtained from 2D orthogonal kilovoltage images, matching with the corresponding digitally reconstructed radiographs. Patients did not have fiducials inserted to assist with localization. Interfractional changes in the bladder and rectal volumes were subsequently measured on the cone beam computed tomography images for each fraction and compared to the shift differences between orthogonal kilovoltage and cone beam computed tomography scans. The proportion of treatment fractions with a shift difference exceeding the planning target volume of 7 mm, between orthogonal kilovoltage and cone beam computed tomography scans, was calculated. RESULTS: The mean vertical, lateral, and longitudinal shifts resulted from 2D match between orthogonal kilovoltage images and corresponding digitally reconstructed radiographs were 0.353 cm (interquartile range: 0.1-0.5), 0.346 cm (interquartile range: 0.1-0.5), and 0.289 cm (interquartile range: 0.1-0.4), compared to 0.388 cm (interquartile range: 0.1-0.5), 0.342 cm (interquartile range: 0.1-0.5), and 0.291 cm (interquartile range: 0.1-0.4) obtained from 3D match between cone beam computed tomography and planning computed tomography scan, respectively. Our results show a significant difference between the kilovoltage and cone beam computed tomography shifts in the anterior-posterior direction ( P = .01). The proportion of treatment fractions in which the differences in kilovoltage and cone beam computed tomography shifts between exceeded the 7 mm planning target volume margin was 6%, 2%, and 3% in the anterior-posterior, lateral, and superior-inferior directions, respectively. CONCLUSION: We prospectively demonstrated that the daily use of volumetric cone beam computed tomography for treatment localization in post-radical prostatectomy patients demonstrated an increased need for a shift in patient position. This suggests that in post-radical prostatectomy patients the daily cone beam computed tomography imaging improved localization of the prostate bed and may have prevented a limited number of geographic misses, compared to daily kilovoltage imaging that was not assisted with fiducials.

What Is the Optimal Radiation Technique for Esophageal Cancer? A Dosimetric Comparison of Four Techniques.

Background Esophageal cancer treatment requires large radiation fields due to the deep location of the esophagus in the mediastinum and the high incidence of radial spread. There is no optimal radiation technique to ensure appropriate target coverage and minimal dose to all normal structures. Methods Fifteen consecutive cases of locally advanced esophageal cancer treated with radical chemoradiation (CRT) were analyzed. The total prescribed dose was 50.4 Gy in 28 fractions. A total of 60 plans were generated for analysis, including four different methods for each case. Method 1 consisted of a four-field conformal technique; method 2 was a two-plan technique (antero-posterior (AP), postero-anterior (PA), two posterior oblique fields (RPO and LPO)); method 3 was a three-field conformal technique (AP, LPO, RPO); and method 4 was a volumetric modulated arc radiotherapy (VMAT) technique. Dose ratios were calculated using the minimum, maximum, mean, and median doses of methods 2-4 over the dose of method 1. Ratios for the planning target volume (PTV) and to surrounding organs were analyzed. Results The mean PTV dose ratio ranged from 0.994 to 1.048 (SD = 0.01) representing an adequate target coverage for all techniques based on an analysis of variance (ANOVA). For the lungs, method 2 had the lowest lung V20 with a ratio of 0.861 (SD = 0.12), whereas method 3 had the highest with 1.644 (SD = 0.14). For the heart, method 3 had the lowest heart V40 with a mean dose ratio of 0.807 (SD = 0.09), whereas method 2 had the highest with 1.160 (SD = 0.11). For the liver, method 2 had the lowest V30 with a mean ratio of 0.857 (SD = 0.1) whereas method 4 had the highest with 1.672 (SD = 0.48). For the spinal cord, method 3 had the lowest mean dose ratio of 0.559 (SD = 0.09) whereas method 2 had the highest with 1.094 (SD = 0.04). Conclusion The four radiation techniques for esophageal cancer treatment were appropriate for target coverage. Method 2 had the most organ-sparing effect for the lungs and liver, and method 3 for the heart and spinal cord. VMAT did not add any significant sparing. A case-by-case decision should be made based on the patient's comorbidities.

Abscopal Effect in a Stage IV Melanoma Patient who Progressed on Pembrolizumab.

In this case report, we present the clinical course of a woman with locally advanced mucosal melanoma of the oral cavity. She was initially treated with surgery with adjuvant local radiation of 50 Gy in 20 fractions. She quickly relapsed with an aggressive regional recurrence of the disease on the neck and with numerous pulmonary metastases. Immunotherapy with pembrolizumab was started, with initial good response and reduction in the size of the lesion in the neck. The regression, however, was short-lived, as the mass quickly grew at a remarkable rate and the lung lesions progressed significantly. Palliative local radiation of 24 Gy in three fractions delivered at days zero, seven, and 21 to the neck mass was eventually given with the goal to alleviate symptoms. An immediate tumor regression was observed after the first fraction of radiotherapy. Remarkably, the lung lesions had also started regressing following radiation. We believe this to be a case of abscopal effect witnessed during the delivery of radiotherapy. A review of the recent literature is also presented here.

Contrast sensitivity of MT receptive field centers and surrounds.

Neurons throughout the visual system have receptive fields with both excitatory and suppressive components. The latter are responsible for a phenomenon known as surround suppression, in which responses decrease as a stimulus is extended beyond a certain size. Previous work has shown that surround suppression in the primary visual cortex depends strongly on stimulus contrast. Such complex center-surround interactions are thought to relate to a variety of functions, although little is known about how they affect responses in the extrastriate visual cortex. We have therefore examined the interaction of center and surround in the middle temporal (MT) area of the macaque (Macaca mulatta) extrastriate cortex by recording neuronal responses to stimuli of different sizes and contrasts. Our findings indicate that surround suppression in MT is highly contrast dependent, with the strongest suppression emerging unexpectedly at intermediate stimulus contrasts. These results can be explained by a simple model that takes into account the nonlinear contrast sensitivity of the neurons that provide input to MT. The model also provides a qualitative link to previous reports of a topographic organization of area MT based on clusters of neurons with differing surround suppression strength. We show that this organization can be detected in the gamma-band local field potentials (LFPs) and that the model parameters can predict the contrast sensitivity of these LFP responses. Overall our results show that surround suppression in area MT is far more common than previously suspected, highlighting the potential functional importance of the accumulation of nonlinearities along the dorsal visual pathway.

Perceptual and neural consequences of rapid motion adaptation.

Nervous systems adapt to the prevailing sensory environment, and the consequences of this adaptation can be observed in the responses of single neurons and in perception. Given the variety of timescales underlying events in the natural world, determining the temporal characteristics of adaptation is important to understanding how perception adjusts to its sensory environment. Previous work has shown that neural adaptation can occur on a timescale of milliseconds, but perceptual adaptation has generally been studied over relatively long timescales, typically on the order of seconds. This disparity raises important questions. Can perceptual adaptation be observed at brief, functionally relevant timescales? And if so, how do its properties relate to the rapid adaptation seen in cortical neurons? We address these questions in the context of visual motion processing, a perceptual modality characterized by rapid temporal dynamics. We demonstrate objectively that 25 ms of motion adaptation is sufficient to generate a motion aftereffect, an illusory sensation of movement experienced when a moving stimulus is replaced by a stationary pattern. This rapid adaptation occurs regardless of whether the adapting motion is perceived. In neurophysiological recordings from the middle temporal area of primate visual cortex, we find that brief motion adaptation evokes direction-selective responses to subsequently presented stationary stimuli. A simple model shows that these neural responses can explain the consequences of rapid perceptual adaptation. Overall, we show that the motion aftereffect is not merely an intriguing perceptual illusion, but rather a reflection of rapid neural and perceptual processes that can occur essentially every time we experience motion.

The role of V1 surround suppression in MT motion integration.

Neurons in the primate extrastriate cortex are highly selective for complex stimulus features such as faces, objects, and motion patterns. One explanation for this selectivity is that neurons in these areas carry out sophisticated computations on the outputs of lower-level areas such as primary visual cortex (V1), where neuronal selectivity is often modeled in terms of linear spatiotemporal filters. However, it has long been known that such simple V1 models are incomplete because they fail to capture important nonlinearities that can substantially alter neuronal selectivity for specific stimulus features. Thus a key step in understanding the function of higher cortical areas is the development of realistic models of their V1 inputs. We have addressed this issue by constructing a computational model of the V1 neurons that provide the strongest input to extrastriate cortical middle temporal (MT) area. We find that a modest elaboration to the standard model of V1 direction selectivity generates model neurons with strong end-stopping, a property that is also found in the V1 layers that provide input to MT. With this computational feature in place, the seemingly complex properties of MT neurons can be simulated by assuming that they perform a simple nonlinear summation of their inputs. The resulting model, which has a very small number of free parameters, can simulate many of the diverse properties of MT neurons. In particular, we simulate the invariance of MT tuning curves to the orientation and length of tilted bar stimuli, as well as the accompanying temporal dynamics. We also show how this property relates to the continuum from component to pattern selectivity observed when MT neurons are tested with plaids. Finally, we confirm several key predictions of the model by recording from MT neurons in the alert macaque monkey. Overall our results demonstrate that many of the seemingly complex computations carried out by high-level cortical neurons can in principle be understood by examining the properties of their inputs.

Pattern motion selectivity of spiking outputs and local field potentials in macaque visual cortex.

The dorsal pathway of the primate visual cortex is involved in the processing of motion signals that are useful for perception and behavior. Along this pathway, motion information is first measured by the primary visual cortex (V1), which sends specialized projections to extrastriate regions such as the middle temporal area (MT). Previous work with plaid stimuli has shown that most V1 neurons respond to the individual components of moving stimuli, whereas some MT neurons are capable of estimating the global motion of the pattern. In this work, we show that the majority of neurons in the medial superior temporal area (MST), which receives input from MT, have this pattern-selective property. Interestingly, the local field potentials (LFPs) measured simultaneously with the spikes often exhibit properties similar to that of the presumptive feedforward input to each area: in the high-gamma frequency band, the LFPs in MST are as component selective as the spiking outputs of MT, and MT LFPs have plaid responses that are similar to the spiking outputs of V1. In the lower LFP frequency bands (beta and low gamma), component selectivity is very common, and pattern selectivity is almost entirely absent in both MT and MST. Together, these results suggest a surprisingly strong link between the sensory tuning of cortical LFPs and afferent inputs, with important implications for the interpretation of imaging studies and for models of cortical function.

LC-MS analysis of antifouling agent Irgarol 1051 and its decyclopropylated degradation product in seawater from marinas in Hong Kong.

A method for the trace analysis of antifouling agent Irgarol 1051 and its decyclopropylated degradation product in seawater was developed by using reversed-phase LC-MS with quadrupole time-of-flight mass spectrometry detection. Atrazine-d(5) was used as the internal standard for the LC-MS determination. Irgarol 1051 and the major degradation product were detected at trace levels in seawater samples collected from several marinas in Hong Kong. The water samples were pre-concentrated by solid-phase extraction with a polystyrene-divinylbenzene cartridge. Quadrupole time-of-flight tandem mass spectrometry in positive ion electrospray ionization was used for the identification and quantitation. The recoveries of Irgarol 1051 and degradation product were >85%. Method detection limits were 0.2ngl(-1) and 0.8ngl(-1) for the antifouling agent and the decyclopropylated degradation product, respectively.