Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Elastomeric biomaterials for tissue engineering

Purpose: To evaluate and confirm efficacy and safety of electrochemotherapy with bleomycin or cisplatin on cutaneous and subcutaneous tumour nodules of patients with malignant melanoma and other malignancies in a multicenter study. Patients and methods: This was a two year long prospective non-randomised study on 41 patients evaluable for response to treatment and 61 evaluable for toxicity. Four cancer centers enrolled patients with progressive cutaneous and subcutaneous metastases of any histologically proven cancer. The skin lesions were treated by electrochemotherapy, using application of electric pulses to the tumours for increased bleomycin or cisplatin delivery into tumour cells. The treatment was performed using intravenous or intratumoural drug injection, followed by application of electric pulses generated by a Cliniporator TM using plate or needle electrodes. Tumour response to electrochemotherapy as well as possible sideeffects with respect to the treatment approach, tumour histology and location of the tumour nodules and electrode type were evaluated. Results: An objective response rate of 85% (73.7% complete response rate) was achieved on the electrochemotherapy treated tumour nodules, regardless of tumour histology, and drug used or route of its administration. At 150 days after the treatment (median follow up was 133 days and range 60‐380 days) local tumour control rate for electrochemotherapy was 88% with bleomycin given intravenously, 73% with bleomycin given intratumourally and 75% with cisplatin given intratumourally, demonstrating that all three approaches were

/pdf/electrochemotherapy-an-easy-highly-effective-and-safe-4cu4r9qcmq.pdf

Electrochemotherapy – An easy, highly effective and safe treatment of cutaneous and subcutaneous metastases: Results of ESOPE (European Standard Operating Procedures of Electrochemotherapy) study

When high-amplitude, short-duration pulsed electric fields are applied to cells and tissues, the permeability of the cell membranes and tissue is increased. This increase in permeability is currently explained by the temporary appearance of aqueous pores within the cell membrane, a phenomenon termed electroporation. During the past four decades, advances in fundamental and experimental electroporation research have allowed for the translation of electroporation-based technologies to the clinic. In this review, we describe the theory and current applications of electroporation in medicine and then discuss current challenges in electroporation research and barriers to a more extensive spread of these clinical applications.

/pdf/electroporation-based-technologies-for-medicine-principles-2dek202ypv.pdf

Electroporation-Based Technologies for Medicine: Principles, Applications, and Challenges

This paper reports results of in vivo experiments that confirm the feasibility of a new minimally invasive method for tissue ablation, irreversible electroporation (IRE). Electroporation is the generation of a destabilizing electric potential across biological membranes that causes the formation of nanoscale defects in the lipid bilayer. In IRE, these defects are permanent and lead to cell death. This paper builds on our earlier theoretical work and demonstrates that IRE can become an effective method for nonthermal tissue ablation requiring no drugs. To test the capability of IRE pulses to ablate tissue in a controlled fashion, we subjected the livers of male Sprague-Dawley rats to a single 20-ms-long square pulse of 1000 V/cm, which calculations had predicted would cause nonthermal IRE. Three hours after the pulse, treated areas in perfusion-fixed livers exhibited microvascular occlusion, endothelial cell necrosis, and diapedeses, resulting in ischemic damage to parenchyma and massive pooling of erythrocytes in sinusoids. However, large blood vessel architecture was preserved. Hepatocytes displayed blurred cell borders, pale eosinophilic cytoplasm, variable pyknosis and vacuolar degeneration. Mathematical analysis indicates that this damage was primarily nonthermal in nature and that sharp borders between affected and unaffected regions corresponded to electric fields of 300-500 V/cm.

In vivo results of a new focal tissue ablation technique: irreversible electroporation

Permeabilization, when observed on a tissue level, is a dynamic process resulting from changes in membrane permeability when exposing biological cells to external electric field (E). In this paper we present a sequential finite element model of E distribution in tissue which considers local changes in tissue conductivity due to permeabilization. These changes affect the pattern of the field distribution during the high voltage pulse application. The presented model consists of a sequence of static models (steps), which describe E distribution at discrete time intervals during tissue permeabilization and in this way present the dynamics of electropermeabilization. The tissue conductivity for each static model in a sequence is determined based on E distribution from the previous step by considering a sigmoid dependency between specific conductivity and E intensity. Such a dependency was determined by parameter estimation on a set of current measurements, obtained by in vivo experiments. Another set of measurements was used for model validation. All experiments were performed on rabbit liver tissue with inserted needle electrodes. Model validation was carried out in four different ways: 1) by comparing reversibly permeabilized tissue computed by the model and the reversibly permeabilized area of tissue as obtained in the experiments; 2) by comparing the area of irreversibly permeabilized tissue computed by the model and the area where tissue necrosis was observed in experiments; 3) through the comparison of total current at the end of pulse and computed current in the last step of sequential electropermeabilization model; 4) by comparing total current during the first pulse and current computed in consecutive steps of a modeling sequence. The presented permeabilization model presents the first approach of describing the course of permeabilization on tissue level. Despite some approximations (ohmic tissue behavior) the model can predict the permeabilized volume of tissue, when exposed to electrical treatment. Therefore, the most important contribution and novelty of the model is its potentiality to be used as a tool for determining parameters for effective tissue permeabilization.

/pdf/sequential-finite-element-model-of-tissue-3ulg2kvobg.pdf

Sequential finite element model of tissue electropermeabilization

One of the ways to potentiate antitumor effectiveness of chemotherapeutic drugs is by local application of short intense electric pulses. This causes an increase of the cell membrane permeability and is called electropermeabilization. In order to study the course of tissue permeabilization of a subcutaneous tumor in small animals, a mathematical model was built with the commercial program EMAS, which uses the finite element method. The model is based on the tissue specific conductivity values found in literature, experimentally determined electric field threshold values of reversible and irreversible tissue permeabilization, and conductivity changes in the tissues. The results obtained with the model were then compared to experimental results from the treatment of subcutaneous tumors in mice and a good agreement was obtained. Our results and the reversible and irreversible thresholds used coincide well with the effectiveness of the electrochemotherapy in real tumors where experiments show antitumor effectiveness for amplitudes higher than 900 V/cm ratio and pronounced antitumor effects at 1300 V/cm ratio.

/pdf/the-course-of-tissue-permeabilization-studied-on-a-39cyhvt6p2.pdf

The course of tissue permeabilization studied on a mathematical model of a subcutaneous tumor in small animals

http://lbk.fe.uni-lj.si/pdfs/bb2007dc.pdf

Real time electroporation control for accurate and safe in vivo non-viral gene therapy

Sequential model of liver tissue electropermeabilisation around two needle electrodes was designed by computing electric field (E) distribution by means of the finite element (FE) method. Sequential model consists of a sequence of static FE models which represent E distribution during tissue permeabilisation. In the model an S-shaped dependency between specific conductivity and E was assumed. Parameter estimation of S-shaped dependency was performed on a set of current measurements obtained by in vivo experiments. Another set of in vivo measurements was used for model validation. Model validation was carried out in three different ways by comparing experimental measurements and modelled results. The model validation showed good agreement between modelled and measured results. The model also provided means for better understanding processes that occur during permeabilisation. Based on the model, the permeabilised volume of tissue exposed to electrical treatment can be predicted. Therefore, the most important contribution of the model is its potential to be used as a tool for determining the electrode position and pulse amplitude needed for effective tissue permeabilisation.

Sequential Finite Element Model of Tissue Electropermeabilisation

Several wound healing rate measures have been introduced with the main goal of enabling quantification of the effects of various therapeutic modalities on the healing of open wounds Different definitions of wound healing rate render comparison of clinical results difficult The goal of the present study was to propose a measure of wound healing rate that is independent of initial wound extent and to present a method of wound healing rate prediction Comparisons were made of wound healing rate defined as absolute area healed per day, percentage of initial area healed per day and advance of the wound margin towards the wound centre per day Analysis was performed on 300 wound cases A disadvantage of wound healing measures that either use absolute area healed per day or percentage of initial area healed per day is their very limited use for comparing healing rates of wounds with different initial sizes This disadvantage was overcome by incorporating a wound perimeter; thus obtaining a measure of the advance of the wound margin towards the wound centre A definition of healing rate expressed as the greatest average wound margin distance from the wound centre divided by the time to complete wound closure is proposed Because not all wounds are closed in the observation period, the time to complete wound closure has to be predicted A method of wound healing rate prediction is presented based on a delayed exponential model the parameters of which are obtained from at least five weekly wound area measurements Paired t-tests between actual time needed to complete wound closure and the predicted time resulted in p = 0062 after four, 0484 after five and 0900 after six weeks of observation

/pdf/a-reliable-method-of-determining-wound-healing-rate-24npwy0qk2.pdf

D. Cukjati

Papers

Sequential finite element model of tissue electropermeabilization

The course of tissue permeabilization studied on a mathematical model of a subcutaneous tumor in small animals

Real time electroporation control for accurate and safe in vivo non-viral gene therapy

Sequential Finite Element Model of Tissue Electropermeabilisation

A reliable method of determining wound healing rate